In 1993, 688,000 horses were estimated to be maintained on tall fescue (Lolium arundinaceum) pasture in the United States (Hoveland, 1993). Many of the dangers associated with the consumption of endophyte-infected fescue have been well described in horses and cattle. In horses, deleterious reproductive effects are the most readily recognized symptoms (Cross et al., 1995). Cattle that graze on endophyte-infected fescue suffer from increased body temperature and a condition known as fescue foot; both of these conditions have been attributed to peripheral vasoconstriction (Strickland et al., 1993). This vasoconstriction results in necrosis of the peripheral tissues of the animal (e.g., feet, tail, and ears; Cross et al., 1995). Although fescue foot has not been reported in horses, loline and ergot alkaloids, such as ergovaline, that are found in endophyte-infected fescue have vasoconstrictive effects on equine vascular tissue in vitro (Klotz and MCDowell, 2010).
Circulation is critical in the hoof of the horse. Blood is essential not only to nourish the digital tissues (Robinson, 1998; Harris et al., 2006) and remove metabolites but also to thermoregulate the hoof (Robinson, 1998). Because of the rigid nature of the hoof wall, the equine foot is particularly susceptible to changes in pressure, with little room for expansion to accommodate edema. Any disruptions to normal circulation may create soundness problems, and lameness has been cited as the most common cause of lost performance (Jeffcott et al., 1982). Because alkaloid compounds associated with endophyte-infected fescue cause vasoconstriction (Strickland et al., 1993; Klotz et al., 2007a) in equine vascular tissue (Klotz and McDowell, 2010), the objectives of this study were to determine if endophyte-infected fescue consumption with ergovaline concentrations similar to those present in pasture alters digital circulation in vivo and is associated with increased lameness in the horse.
MATERIALS AND METHODS
All procedures were approved by the Institutional Animal Care and Use Committee at Kansas State University.
Twelve 3-yr-old American Quarter Horses, 6 geldings and 6 mares, with a mean initial BW of 459 ± 31 kg, were blocked by BW, gender, and hyperkalemic periodic paralysis (HYPP) status [homozygous normal (NN), n = 8; heterozygotes (NH), n = 4] and were assigned to 2 treatment groups: high-endophyte diet (E+, n = 6) and low-endophyte diet (E-, n = 6). Treatment groups were housed in separate drylot pens, each pen measuring approximately 45 × 20 m. All horses had their hooves trimmed by a professional farrier every 6 wk. Horses on each treatment group were trimmed on the same days, and horses were not shod. Horses were maintained on a routine vaccination and deworming program.
Before the study was initiated, horses underwent a 4-wk adaptation period, during which they were accustomed to being caught, tied, and fed the concentrate portion of their ration individually twice daily at 0700 and 1700 h. Horses were gradually introduced to the concentrate mixture that was used in the treatment diets, with all horses receiving endophyte-free fescue seed during the adaptation period. During this period, the quantity of molasses added was adjusted to ensure that all horses consumed the entire concentrate mixture without sorting of the fescue seed. The study began in June and concluded in October. All horses were provided unlimited access to water and a trace mineral salt block throughout the study. The experiment began on d 0, and the 90-d trial was divided into three 30-d segments. On d 0 and at the conclusion of each 30-d period, clinical lameness exams were conducted, horses were weighed, and digital circulation was evaluated using Doppler ultrasonography (Megas ES FD570A; Biosound Esaote, Inc., Indianapolis, IN) and thermographic imaging (ThermaCAM P45HSV; FLIR Systems, Inc., Pleasant Hill, MO).
From d 1 to 30, all horses were provided unlimited access to large, round bales of native prairie hay and were individually fed concentrate twice daily. The concentrate mixture was composed of 2.01 kg of horse feed (Life Design Compete Horse Feed; Nutrena, Minneapolis, MN), 0.23 kg of either Kentucky 31 Tall Fescue seed (E+; Ferry Morse Seed Co., Fulton, KY) or Fuego Tall Fescue seed (E-; Land O'Lakes, Inc., Fort Dodge, IA; Table 1), and 0.02 kg dry molasses (Purina Mills, Richmond, IN). Kentucky 31 fescue seed was integrated into the concentrate at a rate sufficient to bring daily ergovaline consumption in the E+ diet to a minimum of 200 μg/kg, the same dosage reported to elicit fescue toxicities in cattle, which are reported to be more resistant to fescue toxicosis than horses (Strickland et al., 1993; Blodgett, 2001). This dosage also approximates conservative ergovaline concentrations reported in fescue pastures in the central region of the United States (154 to 506 μg/kg; Peters et al., 1992). Ergovaline content in the E- diet remained at undetectable amounts.
|Item||Ergocryptine, μg/kg||Ergocristine, μg/kg||Ergovaline, μg/kg|
From d 31 to 60, the native prairie hay in each diet was replaced by large, round bales of either a high-endophyte (E+) or low-endophyte (E-) variety of Kentucky 31 fescue hay. Administration of the concentrate portion of the diet did not change from the initial feeding period, thus ensuring that each horse in the E+ treatment group received a minimum of 200 μg/kg of ergovaline. Because hay was group fed and individual hay consumption was not measured, total ergovaline consumption was calculated on the basis of an assumed daily DM consumption of 2.0% of BW. Thus, individual total daily ergovaline consumption, including both the hay and concentrate, was estimated to be 280 μg/kg (E+) and 18 μg/kg (E-).
From d 61 to 90, the fescue hay was removed, and native prairie hay was again offered ad libitum to both groups of horses. Although 0.23 kg of fescue seed continued to be administered in the concentrate portion of the diet, it was fed in ground, rather than whole, form during the final 30 d of the trial. Total daily ergovaline consumption was consistent with the first feeding period, but grinding the fescue seed decreased particle size, increased surface area, and may have altered the bioavailability of ergot alkaloids and decreased the potential for sorting of the fescue seed.
Samples of the fescue hay and fescue seed were analyzed using HPLC by the Veterinary Medicine Diagnostic Laboratory at the University of Missouri (Columbia) to quantify common ergot alkaloids in the feedstuffs used in this experiment. The high-endophyte fescue hay (E+) contained 175 μg/kg of ergovaline, whereas the low-endophyte fescue hay (E-) contained 40 μg/kg of ergovaline (Table 1). The Kentucky 31 seed (E+) had an ergovaline content of 4,200 μg/kg, and ergovaline was not detected in the Fuego Tall Fescue seed (E-).
Baseline measurements of the distal limb of each horse were characterized on either d −1 or 0 utilizing Doppler ultrasonography (Megas ES FD570A; Biosound Esaote, Inc.), thermographic imaging (ThermaCAM P45HSV; FLIR Systems, Inc.), and clinical lameness examinations. This protocol was repeated every 30 d throughout the duration of the trial. During each sampling period, measurements were taken on 2 consecutive days to better manage the logistics involved in coordinating all measurements in the sampling protocol. Three horses from each treatment group were sampled on the first day, and the remaining 6 horses were subjected to the same sampling protocol on the next day. Horses were taken to the Kansas State University Veterinary Medical Teaching Hospital and housed in 3.66 × 3.66 m box stalls overnight before the initiation of sampling to allow for acclimation to the air-conditioned environment. The climate-controlled environment also ensured that all measurements (on all days and at all time points) were collected under similar conditions. The legs of the horses were washed to remove any debris, and the sole and frog of each hoof were cleaned.
On the morning of each examination, horses were fed treatment diets in 5-min intervals to allow consistent measurements between horses in relation to feeding time. Most horses finished their concentrate within 30 min, but sampling did not begin for each horse until 60 min after its morning feeding.
Horses were weighed and blood samples were taken approximately 120 min after the morning meal. Approximately 10 mL of blood was collected via jugular venipuncture, refrigerated overnight at 5°C, and centrifuged at 3,000 × g for 15 min at 5°C, and serum was removed. The serum was frozen and sent on dry ice to the School of Animal Sciences at the Louisiana State University (Baton Rouge) for an equine prolactin assay. Prolactin was measured by a double-antibody RIA (Thompson et al., 1986). Intra- and interassay coefficients of variation were 7% and 12%, respectively. Assay sensitivity was 0.2 ng/mL.
Doppler Ultrasonography. The diameter (cm) of the medial palmar artery in the distal left forelimb of each horse, as well as the velocity of blood flow (m/s) though the same artery, was measured using Doppler ultrasonography (Megas ES FD570A; Biosound Esaote, Inc.). All scans were performed by the same board-certified veterinarian on d 0 and 30, with another board-certified veterinarian performing all scans on d 60 and 90. The medial fetlock of the left forelimb was shaved using electric clippers with a size 50 surgical blade before the first sampling. The shaved area was cleaned with alcohol, and a transmission gel was applied to enhance contact of the handheld probe with the limb before each data collection time point. The probe was used to measure diam. and velocity of blood flow through the medial palmar artery, the same location in which Moore et al. (2008) identified vasoconstriction in horses utilizing Doppler ultrasonography. This procedure was repeated at 30-min intervals for 150 min, starting 60 min after the morning meal. Velocity was recorded on each sampling date, but diam. was recorded only on d 30, 60, and 90. After data were collected, volume of blood flow (mL/s) was calculated by multiplying velocity of blood flow (cm/s) by the cross-sectional area of the medial palmar artery (cm2). Rectal temperature was recorded each time a Doppler ultrasonographic examination was conducted.
Thermographic Imaging. A digital thermographic camera with a resolution of 0.05°C (ThermaCAM P45HSV; FLIR Systems, Inc.) was used to measure surface temperature of the hoof as an indicator of blood perfusion. Horses were removed from their stalls and placed on a flat, dry, clean concrete surface. Legs were placed squarely underneath each horse with front feet parallel to each other, and the camera was positioned 61 cm from the apex of the hoof at ground height. The camera was then used to determine surface temperature approximately 0.5 cm below the hairline at the center of each front hoof. The experiment was designed to collect thermographic data at 15-min intervals for 180 min, then at 30-min intervals for an additional 150 min, with a final measurement 60 min later, encompassing a total time frame of 400 min. Because of time constraints and horse handling difficulties on d 0, 7 serial temperatures were recorded. Ten temperature measurements were recorded for all other sampling dates.
Clinical Lameness Examinations. All lameness examinations were conducted by board-certified equine veterinarians who were blind to treatment. One veterinarian performed the lameness examinations on d 0 and 30, and another performed the examinations on d 60 and 90. Lameness examinations were conducted approximately 400 min after the morning feeding, and any horse that would not allow handling during any phase of the lameness examination was excluded from analysis for that particular time point. Initially, horses were examined in a straight path, trotting directly toward and away from the examiner on a flat concrete surface. Each horse was assessed a lameness score from 0 to 5 according to guidelines [American Association of Equine Practitioners (AAEP), 2005)]. A score of 0 indicated no apparent discomfort, and a score of 5 indicated minimal weight bearing.
The examiner then assessed frog and sole sensitivity. Hoof testers, like those used in a typical clinical setting, were used to apply pressure sequentially around the perimeter of the frog and across the sole in each foreleg to ascertain areas of sensitivity. Reactions to pressure were compared with the baseline reaction of each horse. An overall sensitivity score, again on a 0 to 5 scale, was then assigned to each hoof. Values for hoof sensitivity given on d 0 and 30 were combined into 1 score that included both feet. On d 60 and 90, when horses were easier to handle, hoof sensitivity scores were maintained as individual scores and were not combined, instead recorded as front right (FR) and front left (FL). Finally, each horse was longed clockwise and counterclockwise on a concrete surface. The examiner again noted signs of lameness and assessed lameness scores of 0 to 5 at the walk, trot, and extended trot.
All data were analyzed using SAS (SAS Inst. Inc., Cary, NC). The GLM procedure was used when analyzing data obtained through the use of Doppler ultrasonography, thermographic imaging, and prolactin assay. Data were compared at individual time measurements as well as day averages. The model to analyze Doppler and thermography data included the fixed effects of gender, HYPP status, and fescue treatment with BW and rectal temperature as covariates for each measurement period. Lameness scores were categorized as sound or lame with a numerical score of 0 or 1, and analyzed by a P test.
No differences (P ≥ 0.20) were detected between treatment groups or sexes of horses in either concentration of prolactin in the serum on any of the dates tested or changes in prolactin detected from one sampling date to another (Table 2). On d 30, NH horses had greater prolactin concentrations than NN horses, and from d 30 to 60 and d 30 to 90, NH horses had or tended to have a larger decrease in prolactin concentrations than NN horses (P ≤ 0.09; data not shown). Body weights did not differ between treatment groups on any of the dates tested, but when BW gain from d 0 to 30, 0 to 60, 0 to 90, and 60 to 90 was compared, horses on the E+ diet gained or tended to gain more BW than horses on the E- diet (P ≤ 0.08; Table 2). Mares gained or tended to gain more BW than geldings from d 0 to 90 and 60 to 90, and HYPP-heterozygous horses gained or tended to gain more BW than HYPP-negative horses from d 30 to 60 and 30 to 90 (P ≤ 0.09; data not shown). No differences occurred in body temperature between treatment groups on any of the dates or times tested, but E- horses decreased body temperature, and E+ horses increased body temperature from d 30 to 60 (P ≤ 0.05, Table 2). Geldings had lower body temperatures at d 60 and 90 (P ≤ 0.04; data not shown) more than mares (P ≤ 0.07; data not shown) and tended to decrease temperature from d 30 to 60 and 30 to 90. Although the focus of the experiment was digital circulation and lameness in response to dietary treatment, other factors (gender, BW, and HYPP status) affected measurements of digital circulation. These other factors, although not directly related to the question addressed by the experiment, are reported to provide full disclosure to the reader.
||Body temperature, °C
|0||8.02 ±0.53||8.70 ±0.53||0.40||451.9 ±8.2||438.7 ±8.2||0.28||37.6 ±0.3||38.0 ±0.3||0.38|
|30||5.31 ±1.05||4.35 ±1.05||0.53||454.5 ±8.7||452.2 ±8.7||0.85||37.7 ±0.1||37.7 ±0.1||0.95|
|60||4.00 ±1.35||4.01 ±1.34||1.00||479.2 ±9.2||475.0 ±9.2||0.75||37.6 ±0.1||37.8 ±0.1||0.15|
|90||3.00 ±0.67||1.72 ±0.67||0.20||488.0 ±9.5||487.9 ±9.5||1.00||37.6 ±0.1||37.5 ±0.1||0.46|
|0 to 30||−3.20 ±1.13||−4.01 ±1.13||0.63||2.6 ±2.5||13.5 ±2.5||0.01||0.2 ±0.3||−0.3 ±0.3||0.35|
|0 to 60||−4.18 ±1.34||−4.50 ±1.34||0.87||27.3 ±2.0||36.3 ±2.0||0.01||−0.0 ±0.3||−0.2 ±0.3||0.68|
|0 to 90||−4.94 ±1.04||−7.06 ±1.04||0.20||36.1 ±2.9||49.2 ±2.9||0.01||0.1 ±0.3||−0.4 ±0.3||0.27|
|30 to 60||−1.27 ±1.22||−0.31 ±1.22||0.58||24.7 ±2.9||22.8 ±2.9||0.65||−0.2 ±0.1||0.1 ±0.1||0.05|
|30 to 90||−2.31 ±1.25||−2.63 ±1.25||0.86||33.5 ±3.2||35.8 ±3.2||0.62||−0.1 ±0.1||−0.2 ±0.1||0.59|
|60 to 90||−0.98 ±1.35||−2.30 ±1.34||0.50||8.8 ±1.5||13.0 ±1.5||0.08||0.1 ±0.1||−0.2 ±0.1||0.12|
On d 0, no difference (P = 0.21) was observed between treatment groups in velocity of blood flow through the medial palmar artery (data not shown). No differences (P ≥ 0.15) were detected between gender, BW, or HYPP status in velocity of blood flow (data not shown).
Similarly, no treatment differences (P ≥ 0.25) occurred between groups in diam., velocity of blood flow, or volume of blood flow in the medial palmar artery on d 30 (Table 3). At 60 min after feeding, mares had an increased rate of blood flow compared with geldings (P = 0.04; data not shown). No differences (P ≥ 0.37) occurred between sexes at other time periods for velocity or at any time period for volume of blood flow (data not shown). The NH horses had a greater velocity of blood flow at 90 min after feeding than NN horses (P = 0.05; data not shown). No differences (P ≥ 0.11) were detected between horses on the basis of HYPP status at the other time periods for velocity or at any time for diam. or volume (data not shown). Heavier horses had a faster velocity of blood flow at 60 and 90 min after feeding and when averaged over all times after feeding (P ≤ 0.03; data not shown) and tended to have faster blood flow at 120 min after feeding (P = 0.09; data not shown). There was no relationship between BW and velocity of blood flow at 150 or 180 min after feeding, and there was no relationship between BW and diam. of blood vessel at any time after feeding (data not shown). Heavier horses had greater volume of blood flow at 60 and 120 min after feeding than lighter horses (data not shown); however, there was no relationship between BW and volume of blood flow at 90, 150, or 180 min after feeding (data not shown). When all times after feeding were averaged, heavier horses tended to have a greater volume of blood flow than lighter horses (P = 0.10; data not shown). Body temperature had no effect (P ≥ 0.13) on velocity of blood flow, diam. of the medial palmar artery, or volume of blood flow (data not shown).
|Day||Time, min||Velocity, m/s
|30||60||0.29 ±0.03||0.31 ±0.03||0.65||0.33 ±0.05||0.30 ±0.05||0.79||2.46 ±1.00||2.75 ±0.98||0.85|
|90||0.30 ±0.03||0.33 ±0.03||0.47||0.27 ±0.03||0.26 ±0.03||0.83||2.00 ±0.49||2.02 ±0.49||0.98|
|120||0.27 ±0.03||0.30 ±0.03||0.50||0.24 ±0.02||0.26 ±0.02||0.45||1.12 ±0.31||1.69 ±0.31||0.25|
|150||0.27 ±0.02||0.25 ±0.02||0.57||0.22 ±0.02||0.24 ±0.02||0.40||1.00 ±0.21||1.24 ±0.22||0.45|
|180||0.27 ±0.03||0.23 ±0.03||0.33||0.26 ±0.02||0.26 ±0.02||0.91||1.49 ±0.26||1.33 ±0.26||0.68|
|AVG||0.28 ±0.02||0.28 ±0.02||0.83||0.26 ±0.02||0.27 ±0.02||0.89||1.61 ±0.37||1.82 ±0.38||0.69|
|60||60||0.32 ±0.04||0.28 ±0.04||0.59||0.31 ±0.03||0.31 ±0.03||0.96||2.65 ±0.76||2.39 ±0.76||0.82|
|90||0.41 ±0.08||0.41 ±0.08||0.97||0.33 ±0.03||0.32 ±0.03||0.72||3.85 ±1.17||3.94 ±1.18||0.96|
|120||0.41 ±0.06||0.41 ±0.06||0.98||0.34 ±0.02||0.33 ±0.02||0.76||3.88 ±0.95||3.58 ±0.91||0.83|
|150||0.47 ±0.06||0.42 ±0.07||0.67||0.34 ±0.01||0.35 ±0.02||0.66||4.47 ±0.78||4.80 ±0.80||0.79|
|180||0.37 ±0.06||0.38 ±0.06||0.93||0.35 ±0.01||0.33 ±0.01||0.18||3.73 ±0.53||3.62 ±0.51||0.89|
|AVG||0.39 ±0.06||0.39 ±0.05||0.97||0.33 ±0.02||0.33 ±0.02||0.84||3.60 ±0.76||3.75 ±0.75||0.90|
|90||60||0.40 ±0.06||0.41 ±0.06||0.90||0.35 ±0.01||0.33 ±0.01||0.13||3.84 ±0.68||3.49 ±0.74||0.73|
|90||0.44 ±0.03||0.45 ±0.03||0.86||0.35 ±0.01||0.34 ±0.01||0.50||4.31 ±0.44||3.95 ±0.46||0.57|
|120||0.43 ±0.06||0.44 ±0.06||0.84||0.34 ±0.01||0.37 ±0.01||0.05||3.77 ±0.68||4.94 ±0.76||0.29|
|150||0.41 ±0.07||0.42 ±0.07||0.86||0.38 ±0.01||0.37 ±0.01||0.38||4.66 ±0.83||4.63 ±0.83||0.98|
|180||0.35 ±0.04||0.34 ±0.04||0.95||0.39 ±0.02||0.36 ±0.02||0.25||3.95 ±0.50||3.41 ±0.50||0.47|
|AVG||0.40 ±0.05||0.41 ±0.05||0.93||0.36 ±0.01||0.35 ±0.01||0.32||4.11 ±0.45||4.05 ±0.48||0.92|
At 150 min after feeding, mares had a greater arterial diameter and greater volume of blood flow (P ≤ 0.05; data not shown) than geldings. At the same time period, heavier horses had greater diam. and greater volume of blood flow compared with lighter horses (P ≤ 0.04; data not shown), and NH horses tended to have smaller diam. than noncarriers of HYPP (P = 0.07; data not shown). At 180 min after feeding, heavier horses and horses with higher body temperature had greater diam. (P ≤ 0.04; data not shown) and tended to have greater volume of blood flow (P = 0.06; data not shown) in the medial palmar artery.
On d 90, the mean diam. of the medial palmar artery was smaller in the E- treatment group at 120 min following the morning meal (P = 0.05; Table 3). No difference (P ≥ 0.29) was detected between treatment groups in velocity or volume of blood flow through the medial palmar artery (Table 3). At 90 min after feeding, geldings tended to have greater velocity of blood flow compared with mares (P = 0.09; data not shown). Mares had or tended to have greater arterial diameter than geldings at 60, 120, and 150 min after feeding and when averaged over all time periods (P ≤ 0.06; data not shown). Noncarriers of HYPP tended to have an increased diam. of the medial palmar artery than NH horses at 90 min after feeding (P = 0.06; data not shown). Heavier horses had or tended to have increased velocity of blood flow at 90, 120, and 180 min and when averaged for the day (P ≤ 0.08; data not shown). Heavier horses also had or tended to have increased arterial diam. at 150 min and when averaged (P ≤ 0.10; data not shown). Finally, heavier horses also had or tended to have increased volume of blood flow at 90, 120, 150, and 180 min after feeding and when averaged (P ≤ 0.09; data not shown).
When the differences within horse from d 0 to 30 were compared between treatment groups, no differences (P ≥ 0.12) in velocity of blood flow were detected for treatment, gender, or HYPP status (Table 4). Heavier horses had a greater increase in velocity of blood flow when differences between time period measurements were averaged (P = 0.02; data not shown). Body temperature had no effect (P ≥ 0.35) on the change in blood flow from d 0 to 30 (data not shown). When the changes in velocity of blood flow from d 0 to 60 were compared, no differences (P ≥ 0.19) occurred because of treatment, gender, HYPP status, BW, or body temperature (Table 4). When compared between d 30 and 60, the average change in velocity and diam. did not differ between groups for treatment, gender, HYPP status, BW, or body temperature. When the change in velocity of blood flow, diam. of the medial palmar artery, or volume of blood flow was compared between d 0 and 90, 30 and 90, and 60 to 90, no differences (P ≥ 0.31) were evident because of treatment, gender, or HYPP status. Heavier horses tended to have a greater increase in velocity from d 0 to 90 than lighter horses (P = 0.06; data not shown). Horses with higher body temperatures tended to have less change in diameter of the artery between d 60 and 90 (P = 0.07; data not shown).
|0 to 30||0.09 ±0.02||0.11 ±0.02||0.41||—||—||—||—||—||—|
|0 to 60||0.19 ±0.06||0.22 ±0.06||0.90||—||—||—||—||—||—|
|0 to 90||0.21 ±0.05||0.23 ±0.05||0.74||—||—||—||—||—||—|
|30 to 60||0.11 ±0.05||0.11 ±0.05||0.67||0.06 ±0.04||0.07 ±0.04||0.90||1.99 ±1.06||2.02 ±1.05||0.98|
|30 to 90||0.12 ±0.06||0.13 ±0.06||0.96||0.10 ±0.02||0.10 ±0.02||0.97||2.52 ±0.62||2.44 ±0.65||0.93|
|60 to 90||0.00 ±0.07||0.01 ±0.07||0.93||0.03 ±0.01||0.02 ±0.01||0.77||0.38 ±0.74||0.25 ±0.78||0.91|
Across treatments, velocity of blood flow was affected by day of measurement (P ≤ 0.01; Table 5). Velocity was least on d 0 and intermediate on d 30 (P ≤ 0.05). Velocities of blood flow on d 60 and 90 were similar to each other and greater than velocity on d 0 and 30 (P ≤ 0.05). Day of measurement affected arterial diam. (P ≤ 0.01), which was smaller on d 30 than d 60 and 90, which were similar to each other. Similarly, day affected volume of blood flow (P ≤ 0.01), with similar measurements on d 60 and 90 that were greater than d 30. A time of measurement effect occurred for velocity; blood flow was slowest at 135 min after feeding and fastest at 90 min after feeding (P ≤ 0.01; Table 6). Time of measurement had no effect on arterial diam. or volume of blood flow.
|Day||Velocity, m/s||Diam., cm||Volume, cm3/s|
|30||0.27 ±0.03b||0.26 ±0.01a||1.62 ±0.37a|
|60||0.40 ±0.03c||0.33 ±0.01b||3.79 ±0.37b|
|90||0.41 ±0.03c||0.36 ±0.01b||4.24 ±0.37b|
|Time after feeding, min||Velocity, m/s||Diam., cm||Volume, cm3/s|
|60||0.30 ±0.02ab||0.32 ±0.01a||3.03 ±0.30a|
|90||0.33 ±0.02a||0.31 ±0.01a||3.29 ±0.30a|
|120||0.33 ±0.02a||0.31 ±0.01a||3.28 ±0.30a|
|150||0.33 ±0.02a||0.32 ±0.01a||3.53 ±0.30a|
|180||0.28 ±0.02b||0.32 ±0.01a||2.96 ±0.30a|
On d 0, treatment groups did not differ (P ≥ 0.12) in hoof temperature at any of the measurement intervals, and no differences occurred because of gender, BW, or body temperature (Table 7). At 105 min after feeding, NH horses tended to have higher hoof temperatures than NN horses (P = 0.09; data not shown). Hoof temperatures on d 30 were not different because of treatment, gender, HYPP status, BW, or body temperature (Table 8).
|75||32.7 ±0.5||32.7 ±0.5||0.96|
|90||32.7 ±0.5||32.5 ±0.5||0.79|
|105||33.0 ±0.5||33.0 ±0.5||0.98|
|120||32.8 ±0.7||33.3 ±0.6||0.64|
|135||32.6 ±0.6||32.8 ±0.6||0.86|
|150||32.2 ±0.6||32.8 ±0.6||0.56|
|165||33.0 ±0.6||33.4 ±0.6||0.64|
|Avg||32.7 ±0.5||32.9 ±0.5||0.79|
|Time, min||d 30
|75||30.8 ±1.3||28.3 ±1.3||0.23||30.7 ±1.0||28.2 ±1.0||0.12||32.6 ±0.2||32.8 ±0.3||0.59|
|90||30.8 ±1.4||28.0 ±1.4||0.22||31.3 ±1.3||28.3 ±1.3||0.17||33.0 ±0.3||33.3 ±0.3||0.46|
|105||29.4 ±1.8||28.9 ±1.8||0.84||31.3 ±1.1||30.5 ±1.1||0.62||33.3 ±0.2||33.6 ±0.2||0.20|
|120||30.4 ±1.9||29.3 ±1.9||0.68||31.6 ±0.7||31.2 ±0.7||0.68||33.2 ±0.2||33.8 ±0.2||0.02|
|135||29.4 ±2.0||28.9 ±2.0||0.85||31.4 ±0.6||32.0 ±0.5||0.47||33.2 ±0.3||33.2 ±0.3||0.84|
|150||30.2 ±2.0||29.4 ±2.0||0.77||31.8 ±0.5||32.3 ±0.4||0.42||33.3 ±0.3||33.3 ±0.3||0.93|
|165||30.5 ±1.9||28.8 ±2.0||0.54||31.5 ±0.7||32.1 ±0.7||0.58||33.2 ±0.4||33.5 ±0.4||0.59|
|195||30.4 ±1.6||29.0 ±1.6||0.57||32.4 ±0.3||32.8 ±0.3||0.45||32.8 ±0.2||33.1 ±0.2||0.26|
|225||31.3 ±2.1||28.8 ±1.9||0.42||32.5 ±0.3||32.7 ±0.3||0.48||33.0 ±0.2||32.9 ±0.2||0.92|
|285||32.2 ±2.3||28.5 ±1.6||0.24||32.7 ±0.3||33.0 ±0.3||0.47||33.0 ±0.2||33.2 ±0.2||0.45|
|Avg||30.2 ±1.7||28.9 ±1.7||0.59||31.6 ±0.6||31.5 ±0.6||0.91||32.9 ±0.2||33.2 ±0.2||0.21|
On d 60, no differences were measured between treatment groups in hoof temperature. At 75, 135, and 150 min after feeding, mares had higher hoof temperatures than geldings (P ≤ 0.05; data not shown). At 150 min after feeding, NN horses tended to have higher hoof temperatures than NH horses (P = 0.06; data not shown). Heavier horses had or tended to have higher hoof temperatures at 75, 90, 135, and 150 min after feeding and when all times after feeding were averaged (P ≤ 0.08; data not shown). At 135 and 150 min after feeding, horses with lower body temperature had or tended to have higher hoof temperatures (P ≤ 0.07; data not shown).
Horses consuming the E+ diet had higher hoof temperatures compared with the hoof temperatures in horses consuming the E- diet at 120 min after feeding on d 90 (P = 0.02; Table 8). Hoof temperature tended to be higher in NH horses at 285 min after feeding compared with NN horses (P = 0.07; data not shown). At 135 min after feeding, mares tended to have higher hoof temperatures than geldings (P = 0.09; data not shown). Heavier horses had or tended to have higher hoof temperatures than lighter horses at 105, 120, 135, and 150 min after feeding and when all times after feeding were averaged (P ≤ 0.09; data not shown). At 135, 150, and 195 min after feeding, horses with higher body temperatures tended to have lower hoof temperatures (P ≤ 0.10; data not shown).
There were no differences (P ≥ 0.36) in the change in hoof temperature from d 0 to 30 because of treatment, gender, HYPP status, BW, or body temperature (Table 9). No differences (P ≥ 0.19) in change in hoof temperature occurred from d 0 to 60 or from d 30 to 60 because of treatment, gender, BW, or body temperature (Table 9). Horses that were heterozygous for HYPP tended to have a greater decrease in hoof temperature between d 0 and 60 compared with NN horses (P = 0.08; data not shown). No differences (P ≥ 0.13) in change in hoof temperature were evident from d 0 to 90, 30 to 90, or 60 to 90 because of treatment, gender, HYPP status, or BW. From d 60 to 90, horses with higher body temperatures tended to have a greater decrease in hoof temperature than was observed in horses with lower body temperatures (P = 0.09; data not shown). Regardless of treatment, the day affected hoof temperature. Hoof temperatures recorded on d 30 (28.3 ± 0.7) were lower than the temperatures recorded on d 0 (32.6 ± 0.7), 60 (31.9 ± 0.7), or 90 (33.1 ± 0.7; P ≤ 0.01). Across treatments, no time of measurement effect was observed on hoof temperatures.
|0 to 30||−2.5 ±1.4||−4.0 ±1.4||0.48|
|0 to 60||−1.1 ±0.7||−0.15 ±0.7||0.78|
|0 to 90||0.1 ±0.6||0.1 ±0.6||0.99|
|30 to 60||1.2 ±2.0||2.8 ±2.0||0.61|
|30 to 90||2.7 ±1.8||4.5 ±1.9||0.51|
|60 to 90||1.1 ±0.6||1.6 ±0.6||0.50|
Clinical Lameness Examinations
During the baseline evaluations, no differences occurred between treatment groups when horses were trotting on the straight path, when evaluating sensitivity with hoof testers, or during longeing (P = 1.00); data not shown). All horses were clinically sound on d 0. Similarly, when lameness scores were analyzed on d 30, no differences between treatment groups were observed for any of the lameness tests (P ≥ 0.99; Table 10).
On d 60, when comparing treatment groups, horses on the E+ diet were more sensitive to hoof testers on the FL (P = 0.02) and tended to be more sensitive on both feet (P = 0.06; Table 11). During longeing, the E+ group tended to show more lameness compared with the E- group (P = 0.08). No differences in lameness were detected between treatments during observation of horses on the straight line at a trot (P ≥ 0.99). Trotting on a straight line on d 60, geldings showed more lameness than mares (P = 0.01; data not shown), and NH horses tended to show more lameness than NN horses (P = 0.08; data not shown). Additionally, geldings tended to show more lameness during longeing than mares (P = 0.08; data not shown).
On d 90, horses consuming the E+ diet tended to have more sensitivity to the hoof testers in the FR (P = 0.09; Table 11), and a trend emerged for more animals within this group to exhibit some grade of lameness during longeing compared with the E- group (P = 0.12). No differences were detected during observation of horses on the straight line at a trot.
Although broodmares in late gestation consuming endophyte-infected fescue have decreased prolactin concentrations (Strickland et al., 1993; Cross et al., 1995), results of experiments with nonpregnant horses have indicated that fescue consumption does not alter prolactin concentrations (McCann et al., 1992b; Schultz et al., 2006). Thus, detection of differences in prolactin between treatment groups was not expected in the current experiment. Differences in rectal temperature were neither expected between treatment groups nor observed. Although cattle consuming endophyte-infected fescue suffer from increased rectal temperature (Strickland et al., 1993), this does not generally happen in horses (Cross et al., 1995; Youngblood et al., 2004; Webb et al., 2008).
Although some have reported that consumption of endophyte-infected fescue does not alter BW or growth rate in horses (Redmond et al., 1991; McCann et al., 1992b; Schultz et al., 2006), others have reported greater BW loss or decreased growth rate in horses consuming high-endophyte diets (Cross et al., 1995; Webb et al., 2008). Researchers have attributed BW loss or decreased growth rate in horses consuming endophyte-infected fescue to decreased digestibility of endophyte-infected grass or hay (McCann et al., 1992b; Cross et al., 1995). In this study, horses fed the E+ diet gained more BW than those fed the E- diet. Horses were blocked by BW at the start of the study, but those in the E+ treatment group averaged 439 kg on d 0, and those in the E- treatment group averaged 453 kg. By d 90, both groups averaged 488 kg. Those in the E+ treatment group had a lighter BW at the beginning of the trial; thus, the increased BW gain probably was not related to the endophyte content of the diet, but rather to compensatory gain.
Because fescue toxicosis is associated with decreased peripheral circulation and vasoconstriction in cattle (Rhodes et al., 1991; Klotz et al., 2007a, 2009) and vasoconstriction has been demonstrated in equine vascular tissue when exposed to fescue alkaloids in vitro (Abney et al., 1993; Blodgett, 2001), including in the medial palmar artery (Klotz and McDowell, 2010), a similar trend was expected in the current experiment. When analyzing the data collected via Doppler ultrasonography and thermography, a clear trend toward vasoconstriction or reduced circulation to the distal limb in horses consuming the E+ diet was not observed. Despite evidence that equine veins are more sensitive than arteries to a variety of agonists known to induce vasoconstriction (Peroni et al., 2006), the medial palmar artery constricts in vivo in horses consuming a diet that includes ground endophyte-infected fescue seed. Moore et al. (2008) were able to detect this trend using Doppler ultrasonography, and data obtained by the Doppler ultrasonography are repeatable (Menzies-Gow et al., 2008). Data from Doppler ultrasonography are also sensitive in detecting changes in blood flow to the equine limb, detecting differences of 0.005 mL/min in arterial blood flow (Menzies-Gow and Marr, 2007). Horses on the E+ diet in the current experiment received a minimum of 200 μg/kg of ergovaline, but Moore et al. (2008) provided fescue seed as a percentage of BW, with horses consuming 227 to 416 μg/kg of ergovaline. Moore et al. (2008) detected differences only in those horses consuming ground fescue seed (vs. whole seed), so when providing endophyte-infected fescue seed in the diet, particle size may be important in determining physiological effects. Although the fescue seed was ground with a coffee grinder in the final 30 d of the current experiment, no effort was made to measure particle size.
Researchers have reported that cattle suffering from fescue foot demonstrate decreased temperature at the coronary band (Yates et al., 1979), but decreased hoof temperature was not detected in horses consuming the E+ diet in the current experiment. Although concentrations of ergovaline as modest as 50 μg/kg have been associated with reduced blood flow to the skin of wethers and steers (Rhodes et al., 1991), 200 to 280 μg/kg of ergovaline in the current experimental diet (E+) may have been insufficient to elicit a detectible physiologic response in horses; no response was observed regardless of consuming prairie hay or fescue hay or being offered the fescue seed in ground or whole form.
Serial sampling was conducted with both Doppler ultrasonography and thermography to detect changes in circulation that might be dependent on time of feeding, but no clear time-dependent effects were noted in the current experiment. Hoffmann et al. (2001b) have shown that a normal postprandial response in the horse is an increase in blood flow to the hoof, but Hoffmann et al. (2001a) observed no impact of time of day on a variety of blood flow measurements obtained with Doppler ultrasonography. Baseline measurements were not taken in the current experiment before feeding the horses; thus, only postprandial values were available. In bovine lateral saphenous veins exposed to ergovaline, contraction was maintained for 105 min despite repeated buffer replacement (Klotz et al., 2007b). In steers removed from a high-endophyte diet, blood flow to the coronary band increased dramatically within 8 d (Rhodes et al., 1991). Prolactin concentrations in pregnant mares removed from a high-endophyte fescue pasture recovered within 8 d, surpassing those of mares on an endophyte-free pasture (McCann et al., 1992a). Although the half-life elimination of ergovaline is 56.83 ± 13.48 min (Bony et al., 2001), the physiologic response of the animal to endophyte exposure seems to occur over a period of days, and a short-term physiological effect is not apparent on a minute-by-minute basis. If a short-term effect exists, the concentrations of ergovaline in the current diet were likely insufficient to elicit such responses.
Our hypothesis was that if circulation to the digital limb was compromised as a result of consuming the E+ diet, then soundness would also be compromised. By d 60, horses in the E+ treatment group demonstrated greater incidences of hoof sensitivity and lameness during longeing, but by d 90, these differences were diminished, with only a tendency for horses in the E+ treatment group to show increased hoof sensitivity in the front right hoof. Perhaps the increase in ergovaline content offered by the fescue hay from d 30 to 60 was sufficient to elicit a more physiologic response. Because concurrent differences in hoof temperature and blood flow measurements were not detected on d 60 or 90, results of this experiment do not show a clear relationship between fescue consumption, decreased digital circulation, and lameness. However, on d 60, a clear difference occurred in soundness between treatment groups. Considering the moderate ergovaline dosage included in this experiment, this may affect horses grazing on fescue pasture and may merit removing horses from such forage sources, especially in pastures that might have even greater ergovaline content. Even to casual observers, as the horses in the E+ treatment group walked around their pen during this phase of the experiment, some clearly had compromised soundness. No injuries or other potential causes of lameness were noted in any of the horses. Circulatory differences that were undetected with the methodologies used may have occurred. Alternatively, the increased lameness may be caused by another mechanism altogether. If this is the case, more investigation will be required to elucidate the cause of these differences.
Other Fixed Effects and Covariates
Hyperkalemic periodic paralysis in horses results from a point mutation in a sodium channel gene. Sodium channels in the muscle tissue of affected horses are “leaky” and lead to excitable muscle tissue that may contract involuntarily. This is particularly apparent when potassium concentrations increase (Rudolph et al., 1992). If a horse carried the defective allele for HYPP, it influenced some of the measurements obtained via both Doppler ultrasonography and thermography. Although differences were detected in velocity of blood flow, diam. of the medial palmar artery, and hoof temperature at various time points, no clear trend emerged toward increased or decreased digital circulation based on HYPP status.
Several data points collected throughout the 90-d study indicated that mares experienced greater digital circulation than geldings. Often, mares had increased velocity of blood flow, increased arterial diameter, and increased hoof temperature compared with geldings. This could be due, at least in part, to the vasoactive effects of estrogen. The administration of oral estrogen in women, with and without coronary artery disease, results in increased blood flow velocity (Fogelberg et al., 2006); estrogen mediates vasodilation in resistance arteries (Ingelsson et al., 2008) and pulmonary arteries (Lahm et al., 2008), and estrogen is thought to help protect against Alzheimer's disease in human patients by increasing cerebral blood flow (Genazzani et al., 2007). No effort was made in this study to monitor the estrous cycles of the mares or the profiles of steroid hormones, but cyclical changes in estrogen could explain why mares did not consistently demonstrate increased blood flow at each collection time point.
Body weight also influenced digital circulation. On d 30, 60, and 90, observations at several time points indicated that heavier horses had increased velocity of blood flow, arterial diam., and volume of blood flow than lighter horses. This supports the idea that circulation was increased in the hooves of these horses, and the surface temperature of the hoof increased at several time points on d 60 and 90. This seems to be in direct contrast to what others have reported (i.e., blood flow increasing substantially to non-weight-bearing limbs compared with those bearing weight; Hoffmann et al., 2001a). However, a distinct and important difference exists between a non-weight-bearing limb, where resistance to circulation is virtually removed, and the limb of a lighter horse that is weight bearing. Larger horses, by their very nature, are expected to have a larger arterial diam., which was observed fairly consistently throughout the study. Because diam. is a factor in calculating volume of blood flow, it also follows logically that these horses have increased blood flow and thus increased temperature in the distal limb.
This experiment does not provide clear evidence that the consumption of endophyte-infected fescue will alter circulatory patterns to the distal limb in the horse. Perhaps the concentration of ergot alkaloids provided to the horses in this experiment was not sufficient to elicit a physiological response, although 200 μg/kg is within the range (200 to 700 μg/kg) of a typical fescue pasture known to have detrimental effects in pregnant mares (Putnam et al., 1991). When providing fescue seed to horses, particle size may have an important impact on the toxicity of the seed. On d 60 of this experiment, when horses in the E+ treatment group were receiving endophyte-infected fescue hay and thus increased concentrations of ergot alkaloids in the diet (280 μg/kg), some trends emerged toward increased lameness in horses consuming the E+ diet. The observed lamenesses are especially noteworthy when one considers that most fescue pastures will have ergovaline concentrations of at least 150 μg/kg and may be as high as 1,000 μg/kg. Because decreased digital circulation could not be verified, the underlying cause of this lameness remains unidentified. This lameness was obvious enough and of large enough magnitude that it might be prudent for owners to limit the exposure of their horses to endophyte-infected fescue, especially if pastures have ergovaline concentrations greater than 280 μg/kg.