Erica K. Gee

Faecal Microbiota of Forage-Fed Horses in New Zealand and the Population Dynamics of Microbial Communities following Dietary Change

The effects of abrupt dietary transition on the faecal microbiota of forage-fed horses over a 3-week period were investigated. Yearling Thoroughbred fillies reared as a cohort were exclusively fed on either an ensiled conserved forage-grain diet (“Group A”; n = 6) or pasture (“Group B”; n = 6) for three weeks prior to the study. After the Day 0 faecal samples were collected, horses of Group A were abruptly transitioned to pasture. Both groups continued to graze similar pasture for three weeks, with faecal samples collected at 4-day intervals. DNA was isolated from the faeces and microbial 16S and 18S rRNA gene amplicons were generated and analysed by pyrosequencing. The faecal bacterial communities of both groups of horses were highly diverse (Simpson’s index of diversity >0.8), with differences between the two groups on Day 0 (P<0.017 adjusted for multiple comparisons). There were differences between Groups A and B in the relative abundances of four genera, BF311 (family Bacteroidaceae; P = 0.003), CF231 (family Paraprevotellaceae; P = 0.004), and currently unclassified members within the order Clostridiales (P = 0.003) and within the family Lachnospiraceae (P = 0.006). The bacterial community of Group A horses became similar to Group B within four days of feeding on pasture, whereas the structure of the archaeal community remained constant pre- and post-dietary change. The community structure of the faecal microbiota (bacteria, archaea and ciliate protozoa) of pasture-fed horses was also identified. The initial differences observed appeared to be linked to recent dietary history, with the bacterial community of the forage-fed horses responding rapidly to abrupt dietary change.

A cross-sectional survey of Thoroughbred stud farm management in the North Island of New Zealand

Abstract AIM: To obtain initial baseline data on the management of Thoroughbred stud farms in the North Island of New Zealand. METHODS: Data on the management of Thoroughbred stud farms were collected from a sample of 22 stud farms located in the south Auckland/Waikato region (n=15) and lower North Island (n=7) of New Zealand, using a face-to-face survey. The studmaster provided information on the size, scope and management of the farms during the 2004/2005 breeding season. Analysis was based on the location of the farm and size of the breeding operation (number of resident mares). RESULTS: Effective farm size ranged from 20 to 526 ha and averaged 167 (standard error (SE) 36) and 88 (SE 49) ha in the south Auckland/Waikato and lower North Island areas, respectively. Some farms in the Auckland/Waikato region stood shuttle stallions. The median number of stallions per farm was three (range 0.9), and the median mare-to-stallion ratio was 43 (range 10.250). The farms had a mean of 50 (range 7.180) wet mares and 21 (range 0.100) dry mares. The number of mares per breeding stallion increased with increasing size of breeding operation (p=0.04), being 28 (95% confidence interval (CI) = 10.56) vs 40 (95% CI=16.74) vs 74 (95% CI=44.113) for moderate (≤70 mares), medium (90–199 mares) and large (≥200 mares in total) operations, respectively. Seventy-one percent of farms aimed to breed dry mares early in the breeding season, and used a combination of lights, hormone therapy, and rising plane of nutrition to achieve this. Foaling took place in foaling paddocks monitored using a night foaling attendant (17/22) or with foaling alarms (5/22). At birth, 17/22 studmasters routinely administered antibiotics, 14/22 administered tetanus antitoxin, 9/22 administered an enema to foals, and 2/22 did not routinely administer prophylactic treatments. Weaning occurred at 5 (range 3.7–7) months of age, and foals were confined to a box for 1–2 weeks on 16/22 farms. Weaned foals were drenched with anthelmintics every 7 (range 4–9) weeks, and were fed 2.9 (range 1–6) kg of concentrate feed while at pasture until intensive management associated with preparation of the horses for auction began 13 (range 6–20) weeks before the yearling sales. Eight farms weighed the weanlings, at least monthly, to monitor growth. CONCLUSIONS AND CLINICAL RELEVANCE: The management of Thoroughbred horses was relatively consistent throughout the regions surveyed. Utilisation of breeding stallions tended to be more efficient on the larger stud farms in the south Auckland/Waikato region. Even though foals are grown at pasture they are often provided with large quantities of concentrate feed.

Digestible energy intake, dry matter digestibility and mineral status of grazing New Zealand Thoroughbred yearlings.

Abstract AIM: To measure the nutritive value of pasture in terms of digestible energy intake (DEI) and dry matter (DM) digestibility, and to determine the effect of Ca, P, Cu, Zn, and Se supplementation on growth rate and degree of physeal swelling in Thoroughbred yearlings grazed on pasture. METHODS: Fourteen yearling horses were randomly allocated to 1 of 2 groups and rotationally grazed on a ryegrass/white clover pasture for 7 months. One group was supplemented daily with a mineral mix. Liveweight changes were recorded at fortnightly intervals and pasture mineral composition determined at monthly intervals. The DM intake (DMI) was determined from daily faecal DM outputs divided by the indigestible DM fraction (1 — digestible DM) determined from a digestibility study. The DEI was determined from the difference between the gross energy intake and the gross energy faecal output. The DM, gross energy content, crude protein (CP), soluble carbohydrate, acid detergent fibre (ADF), neutral detergent fibre (NDF), lipid and mineral composition of pasture offered and faeces were analysed and their digestibility or apparent absorption calculated. RESULTS: The DM digestibility of pasture averaged 0.64 while the daily DMI and DEI of a 350 kg yearling, gaining 0.6 kg liveweight/day, averaged 6.9 kg DM/day and 78 MJ DE/day, respectively. The mineral supplement had no significant effect on the growth of yearlings grazing pasture that had a mineral composition of (g/kg DM) Na 0.17, K 32.8, Ca 3.6, P 3.4, Mg 2.1, S 3.0, and (mg/kg DM) Cu 7.9, Fe 177, Mn 83, Zn 28 and Se 0.016. Plasma Se was elevated (e.g. 40-218 ώg/l) by Se supplementation while Cu supplementation had no effect on plasma and liver Cu concentrations. Physeal swelling was observed in all horses, regardless of treatment. CONCLUSION: Good quality ryegrass/white clover pasture that had a DE content of 11.3 MJ/kg DM and adequate mineral composition (as observed in this study, with the possible exception of Se) will provide an adequate intake of nutrients to ensure good growth in yearling horses.

Feeding value of pastures for horses

Abstract The feeding value of fresh pasture grazed in situ is determined by animal performance or productivity and could be relatively easily established for growing and lactating horses. Despite this, there is a lack of published information on the relative feeding value of different pastures and forages grazed by horses in New Zealand and the world. In addition, for adult breeding or non-breeding and young or adult sport or performance horses, the definition of feeding value and its determination remain problematic. Limited information suggests that the feeding value of perennial ryegrass-based pasture in New Zealand for young growing horses is high, and growth rates for Thoroughbred horses fed solely on pasture in New Zealand are similar to those reported from the Northern Hemisphere where grain-based supplements are fed in addition to pasture or other forages. Attempts to assess the ability of fresh pastures to meet the nutrient requirements of horses are hampered by problems associated with determination of feed intake by grazing horses and lack of knowledge of the digestibility and utilisation of digested nutrients, including the relative bio availability of macro- and micro-minerals in pasture. A further challenge for future research is to determine the effect of herbage allowance and grazing behaviour, including pasture species preferences, on voluntary feed intake by grazing horses. Grazing pasture has benefits for equine health and well-being including reduced risk of some nutrition-related disorders and reduced prevalence of stereotypic behaviour. Pastured horses have greater freedom for expression of natural behaviours including social interaction and exercise. However, grazing pasture is also associated with animal health problems, particularly parasitism and diseases related to pasture-associated toxins.

Articular / epiphyseal osteochondrosis in Thoroughbred foals at 5 months of age: influences of growth of the foal and prenatal copper supplementation of the dam.

Abstract AIMS: To determine the influence of copper (Cu) supplementation by injection of mares in late gestation on the frequency and severity of osteochondrosis (OC) lesions in their foals at around 160 days of age. To determine if there was any influence of the concentration of Cu in the liver, growth rate, birthweight, weight at 160 days of age, fatness, sex, or year of birth of the foal on the frequency and severity of OC lesions. To determine the influence of dams age, and sex and birthweight of the foal on the growth rate from birth to 160 days of age, and weight at 160 days of age. METHODS: Thirty-three Thoroughbred foals, born in two consecutive years, were weighed every 2 weeks from birth. The dams had been supplemented with parenteral Cu or saline during late gestation, and the supplementation regimens were different in each year. Foals had liver biopsies harvested at birth for determination of Cu concentration. Pasture samples were collected every 4–8 weeks for analysis of concentration of Cu and zinc (Zn). At 160 days of age, articular cartilage of long bones was examined. Gross lesions were counted and scored, then sawn and radiographed, and processed for histopathology. Lesions were given radiographic scores and histopathological scores. Maximum scores for each lesion were combined to give a total OC score for each joint and each foal. The fatness of 20 foals (10 each from Years 1 and 2) at 160 days of age was determined chemically. RESULTS: Supplementation of dams with Cu had no significant effect on the concentration of Cu in the liver of foals at birth, or on the frequency or severity of lesions in articular cartilage at 160 days of age. The Cu and Zn concentrations of pasture were similar in Years 1 and 2, and were lower than current recommendations. All foals in Year 2, and 9/10 foals in Year 1 had irregularities in cartilage that was confirmed histologically to be indicative of OC. The average number of lesions per foal was 4.7 (SD 1.1) and 5.7 (SD 1.1) in Years 1 and 2, respectively. However, the severity of the lesions was considered mild, and no foals showed any clinical evidence of OC while alive. The number of lesions in the tarsocrural (TC) joint and the TC OC score at 160 days were positively associated with average daily weight gain (ADG) in the previous 4 weeks (p=0.005 and p=0.001, respectively). There was no significant effect of sex, fatness, birth-weight, weight at 160 days of age, or year of birth of the foal on the frequency and severity of OC lesions. CONCLUSIONS: Many of the lesions classified as OC, using classification systems described by other authors, were likely to be normal variations of the process of endochondral ossification. Despite the high frequency of such lesions, they were considered to be of minor significance and none were clinically evident. The distribution of lesions was not typical, and most probably reflected the subtlety of the lesions. These results support the hypothesis that Cu is an over-emphasised factor in the aetiopathogenesis of OC. The relationship between subtle macroscopic lesions and lesions resulting in clinical signs of disease requires further investigation.

Chemical body composition of 20 Thoroughbred foals at 160 days of age, and preliminary investigation of techniques used to predict body fatness.

Abstract AIMS: To determine the chemical body composition of Thoroughbred foals born in two consecutive years, and to investigate several techniques used to predict body composition in foals born in the second year, with particular reference to fat. METHODS: The chemical composition of 20 foals at around 160 days of age, born in two years, was determined. In vivo techniques to predict body composition were assessed in 23 foals born in Year 2, before and after euthanasia; 10 of these foals were used for chemical body composition analysis. Techniques to assess body composition in vivo included liveweight (LW), overall and regional condition scores, ultrasonic fat thickness measurements over the ribs and rump, linear measurements and bioelectrical impedance analysis. Correlations were determined between ultrasonic fat thickness, and bioelectrical impedance analysis, before and after euthanasia. Stepwise regression analysis was used to determine the relationships between in vivo techniques used to assess body composition and the chemical body composition of 10 animals. RESULTS: Foals used for analysis of chemical composition weighed between 220.5 and 260.0 kg before euthanasia. Fat content ranged from 5.5-13.0% of the partial empty body-weight (LW less head, gastrointestinal contents, distal limbs and skin). Fillies had significantly more fat mass and percentage fat than colts (p=0.031 for both measurements). The mean chemical composition of the fat-free partial empty body was 73.2% (SD 0.6) water, 22.7% (SD 0.9) protein and 4.1% (SD 0.4) ash. Most of the variation in the concentration of empty body water was associated with variation in the concentration of fat (p<0.001). The live animal overall condition scores were correlated with fat mass and concentration (p=0.006 and p=0.013, respectively; n=10). Condition score over the rib region was highly correlated with fat mass and fat concentration (p=0.004 and p<0.001; n=10). Ultrasound measurements taken 10 cm cranial to the tailhead and 4 cm from midline, used to assess the thickness of rump fat, were correlated with condition score (p=0.001), and explained 71% of the variation in body fat mass (p=0.002; n=10). Nearly 50% of the variation in fat-free mass and partial empty body water mass were associated with variation in the impedance indices calculated from length and bioelectrical impedance analysis measurements (p=0.023 and p=0.026, respectively; n=10). CONCLUSIONS: At around 42% of expected mature weight, fillies were significantly more fat than colts. Condition scores were correlated with partial empty body fat mass, and there was a trend for higher scores in fillies compared to colts. Much of the variation in water or protein mass of the partial empty body could be explained by variations in LW. CLINICAL RELEVANCE: Measurements of LW, rump fat and condition score are useful predictors of the chemical composition of foals at 5 months of age.

Gestation length and occurrence of daytime foaling of Standardbred mares on two stud farms in New Zealand

Abstract AIMS: To examine the gestation lengths and occurrence of daytime foaling of Standardbred mares foaling outdoors at stud farms in Southland, New Zealand (latitude 45˚S). METHODS: Data were collected prospectively at two commercial Standardbred stud farms (Farms A and B), during the 2008/9 and 2009/10 breeding seasons (October to February). For each foaling, the identity of the mare, date and time of foaling, gender of foal, time the mare passed the fetal membranes, time the foal stood, and foaling problems including dystocias, were recorded. The effect of farm, season, gender of foal, month of artificial insemination (AI) or foaling, age and parity of mare on gestation length, percentage of mares foaling during daylight hours, and percentage of foalings recorded as dystocia, were examined. RESULTS: A total of 614 foaling records were obtained from 507 mares. For 594 foalings with complete records, mean gestation length was 349 (SE 0.5) days. Mean gestation length was shorter for fillies (347.8 (SE 0.6) days) than colts (350.3 (SE 0.6) days) p = 0.021) and decreased with month of AI, from October to February (p = 0.001). The time of foaling was bimodally distributed with the primary peak around 0200 hours and a secondary smaller peak around 1300 hours. The percentage of mares foaling in daylight was lower on Farm A (69/285 (24%)) than Farm B (128/313 (41%)) (p = 0.001). Colt foals were less likely to be born during the day than fillies (OR = 0.63; (95% CI = 0.44–0.88); p = 0.008), but there was no effect of age or parity of mare or month of foaling (p>0.05). CONCLUSIONS AND CLINICAL RELEVANCE: Mean gestation length of mares in this study was longer than that previously reported from other countries. Longer gestation length decreases the time available for these mares to get back into foal in the same season. It is important that managers take this into account and ensure optimum conditions for conception; breeding at the first postpartum oestrus may be essential. There were more mares foaling in daylight hours on one stud than has previously been reported. Under these conditions mares should be monitored for foaling during daytime as well as at night.

Determination of the digestible energy intake and apparent absorption of macroelements in pasture-fed lactating Thoroughbred mares

Abstract AIM: To measure the nutritive value of pasture in terms of digestible energy (DE) intake (DEI) and dry matter (DM) digestibility, and to determine the apparent absorption of macroelements in lactating Thoroughbred mares grazed on pasture. METHODS: DM intake (DMI) and DEI were determined from daily faecal DM output measured in grazing mares, divided by the DM indigestible fraction (1-digestible DM), measured in a digestibility trial using pasture-fed mares. Eight lactating mares and their foals, that had a mean age of 40 days, were grazed separately on 50x100 m areas of pasture and daily faecal DM outputs were recorded for 8 days. Five mares and their foals were then placed in individual bare 20x20 m corrals containing custom-made feeding stations for 14 days to determine the indigestible DM fraction. DM, gross energy content, crude protein (CP), soluble carbohydrate, acid detergent fibre (ADF), neutral detergent fibre (NDF), lipid, and macroelement composition of the pasture offered and faeces were determined and their digestibility and/or apparent absorption calculated. RESULTS: DM digestibility of the pasture was 0.6 and the DMI and DEI of a grazing 560 kg mare in early lactation nursing a foal growing at 1.34 kg/day was 13.6 (SE 0.8) kg/day and 146.9 (SE 8.4) MJ DE/day, respectively. Apparent absorptions of the macroelements measured were: Ca 0.75, P 0.43, Mg 0.63, Na 0.78, and K 0.72. CONCLUSIONS: Good quality ryegrass-white clover pasture that had a DE content of 10.8 MJ/kg DM, and a macroelement composition (g/kg) of Ca 3.33, P 3.0, Mg 1.67, Na 1.67, and K 24.2, will provide adequate DMI, DEI, and macroelement intake to lactating Thoroughbred mares.

Sub-optimal efficacy of ivermectin against Parascaris equorum in foals on three Thoroughbred stud farms in the Manawatu region of New Zealand

Abstract AIM: To ascertain whether resistance was present in Parascaris equorum to the macrocyclic lactone anthelmintic ivermectin, using faecal egg count reduction tests. METHODS: Thirty-nine foals aged between 11 and 28 weeks on three Thoroughbred stud farms (Farms A; n=20, B; n=5 and C; n=14) were treated with ivermectin (Day 0) and faecal egg counts (FEC) were monitored before and for 21 (Farms A and B) or 14 (Farm C) days after treatment. On Farms A and B, the foals were treated with a macrocyclic lactone/benzimidazole/praziquantel combination on Day 21 and FEC assessed on Day 35. The three farms were all in the Manawatu region in the southern half of New Zealands North Island. RESULTS: Of the 39 foals, 15 were not shedding P. equorum eggs on the day they were treated with ivermectin, but all 15 did so post-treatment. The FEC on Farms A and B showed no evidence of a reduction at any time between Day 7 and 21; egg output increased steadily over this period. Following combination treatment FEC were reduced by 94% on farm A and 100% on Farm B. On Farm C, there was a 69% reduction in P. equorum FEC on Day 14. Twelve of the 14 foals on Farm C were shedding strongylid eggs on Day 0 and on Day 14 FEC were reduced by 84%. CONCLUSIONS: Without comparison to untreated control animals, these results do not allow a reliable estimation of the exact level of efficacy of ivermectin against P. equorum, but they nevertheless show that, within the farms studied, ivermectin no longer achieved a complete kill of egg-laying adults and that foals treated with this drug may continue to shed considerable numbers of eggs after treatment. The results also indicate that ivermectin had sub-optimal efficacy against strongylid nematodes on one of the farms.

The use of allied health therapies on competition horses in the North Island of New Zealand

Abstract AIM: To obtain data on the use of allied health therapy within competitive equestrian sport in the North Island of New Zealand. METHODS: Data were collected during January 2010 by survey at show jumping and dressage championships in the North Island, and from racing yards in the Central Districts of New Zealand. The survey consisted of 30 open, closed and multiple-choice questions, and was conducted face-to-face, by the same interviewer. Information on the demographics of riders or trainers and horses in each discipline (show jumping, dressage, and Thoroughbred racing), the use of allied health therapy (physiotherapy, chiropractic and equine sports massage) on horses, and knowledge of training and qualifications of the allied health therapists was obtained. Univariable and multivariable logistic regression were used to examine relationships between demographic variables and the use of allied health therapists. RESULTS: In total, 110 riders or trainers participated in the survey. The relative contribution of responses across disciplines was 39/110 (36%), 41/110 (37%) and 30/110 (27%) for show jumping, dressage, and Thoroughbred racing respectively. Allied health therapists were used by 68/110 (62%) respondents to treat their horses. The most common types of allied health therapy used were chiropractic (25/68; 37%) and physiotherapy (16/68; 24%). The main reasons for using allied health therapies were for back pain (22/68; 32%) and lameness (17/68; 25%). Only 5/68 (7%) respondents chose a type of allied health therapy based on veterinary advice, and 49/68 (72%) stated that their veterinarian and allied health therapist did not work together when treating their horses. The final multivariable model for use of allied health therapists included the explanatory variables discipline of the rider or trainer and the number of horses in training per season. CONCLUSIONS AND CLINICAL RELEVANCE: The use of allied health therapies for the treatment of competition and racehorses was widespread. Many riders or trainers perceived allied health therapy to be beneficial, however many therapists and veterinarians do not work together and therefore the integrative treatment approach to rehabilitation is lost.