Lameness resulting from osteoarthritis and other joint diseases is the most common performance-limiting problem in horses.1–3 Common treatments include rest, intra-articular administration of corticosteroids, and administration of NSAIDs and nutritional supplements that purportedly improve joint function.4 Glycosaminoglycans, sodium hyaluronate, and glucosamine chondroitin are among the nutritional supplements most commonly used to improve joint function. Such supplements are widely used both for treatment of lame horses and for prophylaxis in sound horses.5–8 Despite their widespread appeal, however, there is little clinical evidence that these nutritional supplements are efficacious in reducing lameness severity,7,9,10 and many equine veterinarians recommend them more on the basis of a lack of adverse effects than on the basis of any known treatment effect.11
Resveratrol is a natural polyphenolic compound found in many plants, including grape skins.12 In plants, resveratrol functions to protect against fungal infection.13 In mammals, it has anti-oxidative, anti-apoptotic, and anti-inflammatory effects.14–16 The beneficial effects of resveratrol first came to the attention of the medical community after red wine consumption was implicated as a potential explanation for the apparently low incidence of coronary heart disease despite a diet high in saturated fats among the French people (ie, the so-called French paradox).17,18
Anti-inflammatory mechanisms of resveratrol relevant to joint health are inhibition of nuclear factor-κB resulting in suppression of interleukin-1β production and activity, scavenging of reactive oxygen species, and down-regulation of cyclooxygenase-2–regulated pathways.19–23 Because of these properties, resveratrol has been investigated for its anti-osteoarthritic effects in vitro22–26 and in laboratory animals with experimentally induced osteoarthritis21,27 and has been found to have chondroprotective effects. To our knowledge, however, there have been no reported trials of resveratrol's effects in animals or people with naturally occurring joint disease. The purpose of the study reported here, therefore, was to determine the effect of resveratrol administration in performance horses with naturally occurring lameness localized to the distal tarsal joints. We hypothesized that horses treated with resveratrol orally would have greater improvement in lameness severity, compared with change in lameness severity in control horses given a placebo.
Materials and Methods
Study design
The study was designed as a single-center, randomized, blinded, placebo-controlled clinical trial with a simple parallel design, and was performed in compliance with the CONSORT 2010 guidelines (Supplemental Appendix S1, available at avmajournals.avma.org/doi/suppl/10.2460/javma.249.6.650). The study protocol was approved by the Texas A&M University Animal Care and Use and Clinical Research committees. In brief, horses enrolled in the study were randomly allocated to treatment (resveratrol) and control groups. All horses received intra-articular injections of triamcinolone acetonide in the centrodistal and tarsometatarsal joints of both hind limbs and were subsequently fed a supplement containing the assigned treatment for 4 months.
Sample size calculation
A sample size calculation indicated that approximately 20 horses would be needed in each treatment group to detect an increase in the percentage of horses with an improvement in lameness severity from 50% (placebo group) to 90% (resveratrol group), with 80% power and α = 0.05. Because sample size was for the purpose of comparing 2 proportions, an a priori estimate of population variance was not necessary. To account for a 5% loss to follow-up per group, 45 horses were enrolled in the study. Because the treatment was expected to be safe, interim analyses and stopping guidelines were not needed.
Test articles
Active and placebo test articles were prepared by a manufacturer of a commercially available joint supplementa containing resveratrol for horses. Other than the presence of microencapsulated resveratrol,b the ingredients were identical for the active and placebo test articles. Both included fermentation solubles, Saccharomyces cerevisiae 1026, and diatomaceous earth; the active test article contained sufficient resveratrol to result in 1,000 mg of microencapsulated resveratrolb (70% resveratrol and 30% microencapsulant), every 12 hours, when fed as instructed. Both test articles were stored in white plastic containers that had been sealed by the manufacturer. Feeding instructions were affixed to the containers, and containers were identical other than the presence of an additional label with a single number (1 through 45) that designated the group assignment.
Random assignment
Prior to study initiation, the manufacturer of the test article used an online random list generatorc to randomly assign the numbers 1 through 45 to the resveratrol and placebo groups, and containers were labeled as appropriate for the assigned group by the manufacturer. Once labeled, containers were shipped to the Texas A&M University's pharmacy and maintained by the on-site control officer, who was the university pharmacist. The master list of study number assignments was stored in the control officer's locked office. Other than the control officer, study personnel did not have access to the list of study number assignments, and the treating clinician, principal investigator, staff, and owners were not aware of study group assignments until trial completion. When a horse was enrolled, a prescription was submitted to the pharmacy for the next sequential treatment, and the corresponding sealed, numbered container was dispensed to the owner by pharmacy staff, who were also blinded to study group assignment. Twenty-three horses were assigned to the resveratrol group, and 22 were assigned to the placebo group (Figure 1).
Flow diagram for a randomized, controlled trial of the effects of resveratrol administration in performance horses with lameness localized to the distal tarsal joints.
Citation: Journal of the American Veterinary Medical Association 249, 6; 10.2460/javma.249.6.650
Clinical trial enrollment
The study was advertised through various electronic media to owners and referring veterinarians in the university's local area. Horses that were brought to the Equine Lameness and Equine Orthopedics services because of hind limb lameness or poor performance and horses brought to the university specifically for possible enrollment in the study were assessed for study inclusion by the treating clinician. Horses were considered qualified for the study if they were ≥ 3 years old and had a primary hind limb lameness that was localized to the distal tarsal joints by means of diagnostic anesthesia. Horses were excluded if they had received intraarticular medication in any tarsal joint in the previous 6 months, had received NSAIDs in the previous 7 days, were pregnant, had lameness more severe than grade 4 lameness on a scale from 1 to 5,28 or required additional treatments for their lameness, such as shoeing changes or other intra-articular injections.
To determine whether lameness was associated with the distal tarsal joints, horses underwent physical and lameness examinations and diagnostic anesthesia. When distal tarsal joint anesthesia was indicated on the basis of results of the clinical examination, owner permission was sought to test for study qualification. If the owner did not expect he or she would enroll in the trial if the horse were found to be eligible, the horse did not undergo additional inclusion testing (eg, collection of inertial sensor data prior to diagnostic anesthesia).
After owner permission was obtained to test for study qualifications, information was collected on the initial complaint and whether injections of the distal tarsal joints had ever been performed and, if so, when. Subjective and objective lameness evaluations were then performed in a specific pattern. Horses were observed on a loose lead walking in a straight line and trotting on asphalt in a U-shaped pattern consisting of 40 m in a straight line, 20 m in a half circle, and 40 m in a straight line. The pattern was performed both to the left and to the right and was repeated after passive flexion of the distal and proximal hind limb joints, with the half circle performed in the direction of the limb that had been flexed. At the conclusion of this examination, lameness was scored on a scale from 0 to 528 by the treating clinician on the basis of the more severely affected limb. In addition, the attending clinician recorded whether the hind limb lameness was unilateral or bilateral.
Following completion of the lameness examination, objective lameness data were collected for both hind limbs by means of an inertial-based, body-attached sensor system.29,d For collection of sensor data, horses were walked on asphalt in a 40-m straight line with 3 changes of direction (ie, down and back, twice); this resulted in collection of data for approximately 45 strides, depending on stride length. When prompted by the sensor program, outlier strides were not removed by the user. If there was a substantial disruption in the gait during data collection, such as the horse breaking into a canter or bucking, data were discarded and additional data were collected. Sensor data were analyzed by the system as described.29,30 The A1:A2 ratio for each hind limb, which is the measure of vertical pelvic movement versus expected pelvic movement for each stride, and the MAXDIFF and MINDIFF values, which are measures of the degree of asymmetry of pelvis movement over all collected strides, were recorded.
After collection of inertial sensor data, diagnostic anesthesia was performed in the more severely affected hind limb by means of aseptic injection of mepivacainee in the centrodistal and tarsometatarsal joints (3 to 5 mL in each joint). Lameness severity and response to joint flexion were reevaluated 25 to 30 minutes later by observation of the horse at a trot as described, and degree of improvement in the lameness was subjectively determined by the attending clinician. Radiographs were not obtained as part of the study.
Horses that met the inclusion criteria were enrolled in the study with the informed consent of their owners. Enrolled horses were assigned study case numbers in sequence, with the study case numbers corresponding with the numbers on the test articles. For horses enrolled in the study, triamcinolone acetonidef was injected in the centrodistal and tarsometatarsal joints of both hind limbs (4.5 mg in each joint). Horses (n = 4) enrolled by 1 clinician also received 62.5 mg of amikacin in each joint. If synovial fluid was not observed in the hub of the needle, radiography or fluoroscopy was performed to confirm accurate placement of the needle in the joint prior to triamcinolone injection. Following completion of the intra-articular injections, phenylbutazoneg (2 g, IV) was administered, and horses were discharged. Owners were given the test article container and instructed to add 2 scoops to the feed twice daily until a recheck examination 4 months later. Owners were also instructed to administer phenylbutazone (2 g, PO, q 24 h) for 3 to 7 days and to return the horse to full work in 3 to 7 days, with specific intervals varied depending on the attending clinician. Owners were instructed to maintain a form in which they recorded medications or supplements added or ceased during the study period. Otherwise, owners were instructed to not modify the diet or turnout or exercise regimens of enrolled horses from their normal routine and performance schedule.
Follow-up procedures
Follow-up consisted of administration of a questionnaire 2 and 4 months after horses were enrolled in the study and reexamination at the hospital 4 months after study enrollment. A calendar was maintained to send automatic reminders to alert the principal investigator and hospital reception staff of the need for questionnaires and recheck examinations. The principal investigator called the owner or, if different from the owner, rider of the horse to administer the questionnaire by telephone 2 months after study enrollment. The questionnaire asked interviewees to describe any adverse events that could possibly have been related to the study supplement; whether the horse had been eating the supplement well and, if not, how much the horse was consuming; whether the horse had missed any doses of the supplement and, if so, how many; how much and how often the horse was worked and whether the owner considered this to be full work; whether the horse's performance was better, worse, or the same, compared with performance prior to study enrollment; whether the horse was performing below, at, or above expectations; whether the horse was showing any signs of lameness and, if so, which limbs were affected; whether the owner was satisfied with how the horse was doing (yes vs no); and whether there had been any changes in supplements or other medications administered during the study period and, if so, to list them.
When the appointment was made for the 4-month recheck examination, the owner was reminded to not administer any NSAIDs during the 7 days prior to the examination. During the 4-month recheck examination, the same questionnaire was administered by the attending clinician. In addition, the owner was asked whether the horse had received any NSAIDs in the 7 days prior to the examination. If NSAIDs had been administered, the recheck examination and questionnaire were rescheduled for the following week.
After completion of the questionnaire during the 4-month examination, subjective and objective lameness evaluations were performed in the same manner as during the enrollment examination, except that diagnostic anesthesia was not performed. Although not part of the study, owners were offered intra-articular injection of triamcinolone at the conclusion of data collection, and whether this was elected was recorded.
Data analysis
Data were transferred to electronic files at the conclusion of the trial and after unblinding by an individual not involved in study design or data interpretation. The change in A1:A2 ratio was calculated as the difference in the sum of the A1:A2 ratios for the right and left hind limbs for the enrollment and recheck examinations. The change in MAXDIFF and MINDIFF was calculated by determining the absolute difference between enrollment and recheck examination values. When the absolute value for the recheck examination was greater (ie, farther from 0) than the absolute value for the enrollment examination, the absolute change was assigned a negative value, and when the absolute value for the recheck examination was less (ie, closer to 0) than the absolute value for the enrollment examination value, the absolute change was assigned a positive value. When the value changed signs (eg, from negative to positive) between the enrollment and recheck examinations, the absolute change was assigned a positive value if the recheck value indicated the horse was less lame (ie, was closer to 0) and was assigned a negative value if the recheck value indicated the horse was more lame (ie, was farther from 0).
Data were imported into a commercial statistical software programh for variable coding and analysis. The statistician was aware of treatment group assignment. The Fisher exact test was used to compare proportions of favorable owner and rider responses (eg, a response of better, vs worse or the same, when asked how the horse's performance was, compared with performance prior to study enrollment) between the resveratrol and placebo groups at 2 and 4 months. Point estimates and 95% confidence intervals for each risk difference and risk ratio were calculated with standard software.i Similarly, the Fisher exact test was used to compare proportions of horses in the 2 groups with an improvement in subjective lameness score between the enrollment examination and the 4-month recheck examination. The Wilcoxon rank sum test was used to compare differences in inertial sensor values at 0 and 4 months between the resveratrol and placebo groups, following use of the Shapiro-Wilk test to confirm that these data were not normally distributed. For all analyses, values of P < 0.05 were considered significant.
Results
The initial enrollment period of 18 months was extended by 12 months because initial patient enrollment was below expectations. With the additional 12 months, the target of 45 horses was met, and 45 horses were enrolled in the study. Therefore, the study was performed over the 2.5-year period from February 2013 to September 2015. Population characteristics at enrollment, including signalment, athletic use, initial complaint, and previous tarsal joint injections, and time between enrollment and collection of follow-up data did not differ between groups (Table 1).
Population characteristics for performance horses with lameness localized to the distal tarsal joints that received intraarticular injections of triamcinolone in the centrodistal and tarsometatarsal joints of both hind limbs and were then fed a supplement containing resveratrol (n = 21) or a placebo (20) for 4 months.
| Group | ||
|---|---|---|
| Variable | Resveratrol | Placebo |
| Age (y) | 12.4 ± 6.5 | 10.7 ± 6.0 |
| Use | ||
| Dressage | 4 | 4 |
| Eventing | 4 | 3 |
| Jumping | 2 | 3 |
| Western performance | 4 | 7 |
| Pleasure or trail riding | 2 | 1 |
| Western show | 5 | 2 |
| Sex | ||
| Mare | 7 | 5 |
| Gelding | 14 | 14 |
| Stallion | 0 | 1 |
| Breed | ||
| Warmblood | 5 | 5 |
| Quarterhorse or Paint | 10 | 11 |
| Thoroughbred | 6 | 3 |
| Arab | 0 | 1 |
| Initial complaint | ||
| Lameness | 12 | 9 |
| Possible study inclusion | 9 | 11 |
| Previous tarsal joint injection | ||
| Yes | 7 | 5 |
| No | 13 | 14 |
| Unknown | 1 | 1 |
| Time to follow-up evaluations (d) | ||
| 2-mo evaluation | 61 ± 7 | 59 ± 13 |
| 4-mo evaluation | 118 ± 9 | 119 ± 15 |
Data represent mean ± SD or number of horses.
Owners of 3 horses (1 in the resveratrol group and 2 in the placebo group) could not be contacted to complete either questionnaire despite multiple attempts and did not return their horse for the 4-month recheck examination. The owner of 1 horse in the resveratrol group could not complete either questionnaire because the horse sustained an injury to the right forelimb and was on strict stall rest. Although this horse was returned for the 4-month recheck examination, all data were excluded. Therefore, there were 21 resveratrol-treated and 20 placebo-treated horses with complete data for analysis.
Two months after study enrollment, the percentage of riders who reported that the horse's performance was better, compared with worse or the same, was significantly (P = 0.04) higher for the resveratrol group (20/21 [95%]) than for the placebo group (14/20 [70%]; Table 2). Palatability of the supplement (yes vs no), whether the horse had returned to full work (yes vs no), whether signs of lameness were present (yes vs no), performance compared with expectations (at or above vs below), and whether the owner was satisfied with how the horse was doing (yes vs no) did not differ significantly between groups.
Outcome 2 and 4 months after study enrollment for the horses in Table 1.
| Group* | |||||
|---|---|---|---|---|---|
| Variable | Resveratrol | Placebo | P value† | Risk difference (95% CI) | Risk ratio (95% CI) |
| 2 months | |||||
| Horse performance (better vs worse or the same) | 20/21 (95) | 14/20 (70) | 0.04 | 0.25 (0.03 to 0.47) | 1.36 (1.01 to 1.84) |
| Returned to full work (yes vs no) | 9/21 (43) | 9/20 (45) | 1.0 | −0.02 (−0.33 to 0.28) | 0.95 (0.48 to 1.9) |
| Performance compared with expectations (at or above vs below) | 20/21 (95) | 16/20 (80) | 0.2 | 0.15 (−0.05 to 0.35) | 1.19 (0.94 to 1.51) |
| Lame (yes vs no) | 1/21 (5) | 4/20 (20) | 0.2 | −0.15 (−0.35 to 0.05) | 0.24 (0.03 to 1.95) |
| Owner satisfied (yes vs no) | 21/21 (100) | 18/20 (90) | 0.2 | 0.1 (−0.03 to 0.23) | 1.11 (0.96 to 1.29) |
| Palatable (yes vs no) | 20/21 (95) | 19/20 (95) | 1.0 | NA | NA |
| Missed > 7 doses (yes vs no) | 1/21 (5) | 0/20 (0) | 1.0 | NA | NA |
| 4 months | |||||
| Horse performance (better vs worse or the same) | 18/21 (86) | 10/20 (50) | 0.02 | 0.36 (0.09 to 0.62) | 1.71 (1.07 to 2.75) |
| Returned to full work (yes vs no) | 8/21 (38) | 7/20 (35) | 1.0 | 0.03 (−0.26 to 0.33) | 1.09 (0.48 to 2.44) |
| Performance compared with expectations (at or above vs below) | 19/21 (90) | 17/20 (85) | 0.6 | 0.05 (−0.15 to 0.26) | 1.06 (0.85 to 1.34) |
| Lame (yes vs no) | 3/21 (14) | 6/19 (32) | 0.3 | −0.17 (−0.43 to 0.08) | 0.45 (0.13 to 1.56) |
| Owner satisfied (yes vs no) | 18/20 (90) | 14/18 (78) | 0.4 | 0.12 (−0.11 to 0.35) | 1.16 (0.87 to 1.54) |
| Palatable (yes vs no) | 21/21 (100) | 19/20 (95) | 0.5 | NA | NA |
| Missed > 7 doses (yes vs no) | 0/21 (0) | 0/20 (0) | NA | NA | NA |
| Repeated intra-articular injections (yes vs no) | 14/21 (67) | 13/20 (65) | 1.0 | 0.02 (−0.27 to 0.31) | 1.03 (0.66 to 1.59) |
Data represent number of horses with the factor of interest/number of horses evaluated (%).
Fisher exact test.
NA = Not applicable.
Four months after study enrollment, the percentage of riders who reported that the horse's performance was better, compared with worse or the same, was still significantly (P = 0.02) higher for the resveratrol group (18/21 [86%]) than for the placebo group 10/20 [50%]; Table 2). Palatability of the supplement, whether the horse had returned to full work (yes vs no), whether signs of lameness were present (yes vs no), performance compared with expectations (at or above vs below), and whether the owner was satisfied with how the horse was doing (yes vs no) did not differ significantly between groups. The percentage of owners electing follow-up injection of the distal tarsal joints did not differ between treatment groups. No owners reported discontinuing any other joint supplements during the study period. The percentages of owners who reported administering new joint supplements, NSAIDs, or other medications were not significantly different between the groups. The percentage of owners who, at 2 months, reported seeing adverse effects did not differ significantly (P = 1.0) between the resveratrol group (5/21 [24%]; alopecia in the shoulder region, itchy tailhead, sneezing, smells bad, and more energetic than usual) and the placebo group (5/20 [25%]; fidgety in stall, stone bruise, colitis, more aggressive than usual, and increased water content in manure). At 4 months, the percentage of owners who reported seeing adverse effects was also not significantly (P = 0.2) different between the resveratrol group (2/21 [10%]; parks out when standing and excess energy) and the placebo group (5/20 [25%]; stumbling, colitis, white line disease, lost weight, and loose manure).
There was only 1 horse (resveratrol group) that was reported at the 2-month or 4-month questionnaire to have missed > 7 doses of the supplement. The owner reported that this horse was able to reach the container of supplement from its stall and consumed the entire amount of supplement the day after enrollment. The container was replaced with the appropriate supplement 1 week later by the manufacturer in an identical container marked in the same manner.
At the enrollment and recheck examinations, the A1:A2 ratio for horses in the resveratrol group was not significantly different from the ratio for horses in the placebo group (P = 0.1 and 0.9, respectively). Also, the distribution of lameness scores also did not differ significantly between groups at either examination time (Table 3), and the percentage of horses with a 1-grade or more improvement in lameness score at the 4-month recheck examination, compared with the score assigned at the enrollment examination, was not significantly (P = 0.48) different between the resveratrol (5/21 [24%]) and placebo (3/20 [15%]) groups. However, the change in A1:A2 ratio between the enrollment and 4-month recheck examinations was significantly (P = 0.048) better for horses in the resveratrol group (median change in A1:A2 ratio, 0.024; interquartile [25th and 75th percentiles] range, −0.032 to 0.124) than for horses in the placebo group (median change, −0.017; interquartile range, −0.074 to 0.013; Figure 2). The change in MINDIFF between the enrollment and 4-month recheck examinations was not significantly (P = 0.3) different between the resveratrol (median change, 0.9015) and placebo (median change, 0.1425; Figure 3) groups. Similarly, the change in MAXDIFF between the enrollment and 4-month recheck examinations was not significantly (P = 0.14) different between the resveratrol (median change, 1.148) and placebo (median change, −09510) groups.
Subjective lameness scores at the time of the enrollment and 4-month recheck examinations for the horses in Table 1.
| Group | ||
|---|---|---|
| Variable | Resveratrol (n = 21) | Placebo (n = 20) |
| Enrollment examination | ||
| Grade 1 | 1 (5) | 3 (15) |
| Grade 2 | 5 (24) | 8 (40) |
| Grade 3 | 14 (67) | 12 (60) |
| Grade 4 | 1 (5) | 0 (0) |
| Bilateral | 13 (62) | 10 (50) |
| 4-month examination | ||
| Grade 0 | 1 (5) | 2 (10) |
| Grade 1 | 1 (5) | 1 (5) |
| Grade 2 | 4 (19) | 2 (10) |
| Grade 3 | 15 (71) | 15 (75) |
| Bilateral | 13 (62) | 13 (65) |
| Predominant limb switched | 5 (24) | 7 (35) |
| Improved ≥ 1 lameness grade | 5 (24) | 3 (15) |
Data represent number (%) of horses.
Lameness scores were assigned on a scale from 0 to 5.
Scatterplot of the change in hind limb A1:A2 ratio between the enrollment and 4-month recheck examinations for the horses in Figure 1. A positive value represents an improvement in lameness severity, a negative value represents a worsening in lameness severity, and a value of 0 represents no change in lameness severity.
Citation: Journal of the American Veterinary Medical Association 249, 6; 10.2460/javma.249.6.650
Scatterplots of the change in maximum (MAXDIFF; A) and minimum (MINDIFF; B) pelvic height difference between the enrollment and 4-month recheck examinations for the horses in Figures 1 and 2. A positive value represents an improvement in lameness severity, a negative value represents a worsening in lameness severity, and a value of 0 represents no change in lameness severity.
Citation: Journal of the American Veterinary Medical Association 249, 6; 10.2460/javma.249.6.650
Discussion
Results of the present study suggested that in performance horses with lameness localized to the distal tarsal joints, injection of triamcinolone in the centrodistal and tarsometatarsal joints of both hind limbs followed by oral supplementation with resveratrol for 4 months resulted in reduced lameness, compared with triamcinolone injection and supplementation with a placebo. Specifically, the percentage of riders who reported that the horse's performance was better, compared with worse or the same, was significantly higher for the resveratrol group than for the placebo group 2 and 4 months after study enrollment, and the change in A1:A2 ratio between the enrollment and 4-month recheck examinations was significantly better for horses in the resveratrol group than for horses in the placebo group. On the other hand, owner-reported performance compared with expectations (at or above vs below) at 2 and 4 months, owner satisfaction at 2 and 4 months, percentage of horses reported by the owner to have lameness at 2 and 4 months, percentage of horses that had returned to full work at 2 and 4 months, the A1:A2 ratio at 4 months, the distribution of lameness scores at 4 months, the percentage of horses with an improvement in lameness score at 4 months, and the change in MINDIFF and MAXDIFF between the enrollment and 4-month recheck examinations did not differ between groups.
Because all horses in the present study received a treatment (intra-articular administration of triamcinolone) known to be effective for lameness localized to the distal tarsal joints, we did not collect follow-up data until 2 months after enrollment. This 2-month period was chosen on the basis of findings of a retrospective study31 involving 51 horses with lameness associated with the distal tarsal joints. In that study, there was a recurrence of lameness in 90% of treated horses a median of 56 days after intra-articular injection of a corticosteroid. Therefore, we expected 2 months would be the earliest that differences could be detected if they existed. On the basis of results of the same study,31 we performed a final evaluation 4 months after enrollment, because this time period was beyond the expected duration of efficacy for intra-articular triamcinolone injection.
For the present study, we expected that 4 months after study enrollment, approximately 50% of horses that received the placebo would be reported by their riders to be better, and this was in fact what we found. This expectation was based on previously reported long-term success rates following intra-articular injection of 38% (distal tarsal joints) and 50% (multiple joints).31,32
Blinding of both owners and attending clinicians is critically important in randomized clinical trials, especially when subjective end points are used. In the present study, all participants other than the pharmacist who was the on-site control officer were blinded to treatment group assignment. The pharmacist had no direct interactions with the horse owners, attending clinicians, principal investigator, or staff until trial completion. All treatment containers were dispensed by blinded pharmacy staff in sequential order of enrollment. Because unblinding can accidentally occur because of differences in product appearance or labeling,33,34 we ensured that the test articles used in the present study were identical, other than the sequential numbers on the containers. To the best of our knowledge, there were no adverse effects or other signs that could have led to owner identification of treatment group assignment during the study period.
Both 2 and 4 months after study enrollment, the percentage of riders who reported that the horse's performance was better, compared with worse or the same, was significantly higher for the resveratrol group than for the placebo group. This was despite a lack of difference between groups in percentages of horses that were reportedly lame (yes vs no), percentages of owners satisfied with how the horse was doing (yes vs no), and percentages of horse performing up to expectations (at or above vs below expectations). However, we do not believe that was surprising, given that lameness localized to the distal tarsal joints in performance horses is often mild and can be bilateral, with the result that riders often notice poor or reduced performance rather than overt hind limb lameness.35 Because riders are attuned to minor differences in their horse's performance, we believe that rider opinion as to whether the horse's performance was better, compared with worse or the same, was clinically relevant. Had horses had more obvious signs of lameness that were easier for owners to appreciate, such as a unilateral head nod, it is possible that we would have identified differences between groups in regard to these other outcome measures.
Although the inertial sensor system used in the present study revealed a significant difference in the change in A1:A2 ratio between groups in the present study, the distribution of lameness scores and the percentage of horses with an improved subjective lameness score did not differ between groups. We believe this can be attributed to inherent limitations of a subjective scoring system, under which a single lameness grade can cover a broad range of lameness severities. For example, with the subjective scoring system we used, a horse with an obvious, severe lameness consistently seen when the horse was trotted in a straight line on asphalt but not when walked and another horse with a consistent mild lameness just detectable when the horse was trotted on asphalt but not when walked would both receive a score of 3. Thus, clinically relevant change in lameness severity between the enrollment and 4-month recheck examinations could have gone undetected with the subjective scoring system we used.
In contrast to the lack of difference in subjective lameness scores between groups, we found a significant difference in the change in A1:A2 ratio between the enrollment and 4-month recheck examinations. The principal advantages of the inertial sensor system are that the data are repeatable36,37 and consistent over time38 and the system is able to detect subtle lameness and subtle changes in lameness with greater accuracy, compared with subjective evaluation by a veterinarian and force plate analysis.38–40 Thus, the inertial sensor system is especially useful when looking for mild improvements in lameness over time.39
In the present study, we found a significant improvement in the change in the A1:A2 ratio in the resveratrol versus the placebo group but no difference in the change in MAXDIFF or MINDIFF between groups. Lameness localized to the distal tarsal joints is commonly bilateral, and the predominant limb can switch over time.35 Twenty-three of the 41 (56%) horses had bilateral hind limb lameness at the time of enrollment, and in 12 of the 41 (29%), the predominant limb switched from the time of enrollment to the 4-month recheck examination. Both MAXDIFF and MINDIFF are measures of pelvic symmetry. Therefore, horses with bilateral lameness could have artificially low (ie, closer to 0) absolute values for MAXDIFF and MINDIFF, compared with values for horses with unilateral lameness. In contrast, the A1:A2 ratio is not affected when horses alternate between left and right hind limb lameness.41 Therefore, we believe the sum of the left and right hind limb A1:A2 ratios is a better measure for horses with lameness localized to the distal tarsal joints.
The clinical relevance of changes in the A1:A2 ratio41 and changes in MAXDIFF and MINDIFF39 in horses with hind limb lameness has previously been demonstrated. In a study41 involving 17 horses, the mean hind limb A1:A2 ratio was 0.17 when the horses were considered to be sound. By comparison, median change in the A1:A2 ratio between the enrollment and 4-month recheck examinations was 0.024 for horses in the resveratrol group and −0.017 for horses in the placebo group in the present study, and we believe the difference between these values was clinically relevant as well as statistically significant.
The bioavailability of most nutraceuticals following oral administration has been questioned,42 and it is known that microencapsulation is responsible for the bioavailability of resveratrol.43 When not microencapsulated, resveratrol is rapidly metabolized and excreted before the compound can reach the tissues.43 Although we do not have bioavailability data for the microencapsulated resveratrol product we tested, Ememe et al44 reported significant reductions in the serum concentration of malondialdehyde, an antioxidant marker, and the serum activity of glutathione peroxidase for 4 weeks in horses administered resveratrol prepared by the same manufacturer of the product we tested. Because the manufacturing process is important in resveratrol preparation, products from other manufacturers would need to undergo similar testing in lame horses or bioavailability equivalency testing prior to extrapolation of our results.
In the present study, we did not obtain radiographs of the affected joints. Conceivably, tarsal radiography could have been used to stratify horses on the basis of severity of radiographic changes prior to assignment to groups. However, for horses with distal tarsal joint lameness, severity of radiographic abnormalities does not reflect the degree of clinical lameness or the expected response to intra-articular corticosteroid injection.31 A benefit of the randomized clinical trial design is that unknown potential confounders are typically balanced between study groups.
An important limitation of clinical trials involving patients with naturally occurring disease is the potential for misdiagnosis at the time of study enrollment, leading to the inclusion of animals that do not truly have the condition of interest. With reference to the present study, horses were included only if substantial improvement in lameness severity was seen following diagnostic distal tarsal joint anesthesia. It is still possible that horses with proximal suspensory desmitis, for example, rather than lameness from the distal tarsal joints could have been included.31,45 However, because horses were randomly assigned to treatment groups, the effect of this should have been minimized. Only 1 of 42 owners elected to have additional diagnostic testing performed at the time of the 4-month recheck examination because of a lack of improvement, and the horse was found to have proximal suspensory desmitis. This horse was in the resveratrol group, and the rider had reported that the horse's performance was the same on both the 2-month and 4-month questionnaires.
A limitation of the present study was that we do not have evidence that owners appropriately fed the supplement for the 4 months of the study. However, owner reports on palatability and whether doses were missed were not different between groups. One way to test for owner compliance might have been to weigh containers at the recheck examination, but we did not do this. Because of blinding, we believe that owner compliance would not have differed between groups.
Prospective trials in veterinary medicine are uncommon,46 and well-designed randomized clinical trials involving performance horses are especially rare.7 For the present study, we elected to include horses with lameness localized to the distal tarsal joints for several reasons. First, distal tarsal joint lameness is common in performance horses, the treatment is relatively uniform among clinicians, and the expected outcome is fairly well understood. Second, of the many types of lameness in horses, distal tarsal joint lameness is likely to be the most uniform, with distal tarsal joint lameness most likely a result of synovitis or osteoarthritis and not due to osteochondritis dissecans, subchondral cyst-like lesions, osteochondral fragmentation, or tendon or ligament injury. Third, all horses, even those in the placebo group, were given intra-articular injections of triamcinolone, an effective and well-accepted treatment for distal tarsal joint lameness. Other factors that we believe contributed to the success of owner enrollment and high rate of return for follow-up examinations were the relatively short duration of the study (4 months), lack of crossover design, and offering of repeated distal tarsal joint injections at no cost to the owner after completion of the 4-month recheck examination.
There is in vitro22–24,47–51 and in vivo21,27 evidence that resveratrol may be useful in the treatment of osteoarthritis. However, this type of evidence doesn't fully incorporate the multifactorial nature of lameness in performance horses. Long-term, randomized, controlled clinical trials in horses with naturally occurring disease provide clinicians with the most clinically applicable information.10,52
In conclusion, horses in the present study that received resveratrol supplementation for 4 months following intra-articular injection of triamcinolone in the centrodistal and tarsometatarsal joints were significantly less lame on the basis of objective (change in A1:A2 ratio) and subjective (rider assessment of horse performance) criteria than were horses that received triamcinolone injections and a placebo. Thus, results suggested that oral resveratrol supplementation may be efficacious in horses with lameness.
Acknowledgments
Supported by Equithrive.
The authors declare that there were no conflicts of interest; Equithrive did not have any involvement in the study design, data analysis and interpretation, writing and publication of the manuscript, or manuscript approval.
Presented in abstract form at the annual conference of the Veterinary Orthopedic Society, Big Sky, Montana, February–March 2016.
ABBREVIATIONS
| A1 | Lameness amplitude |
| A2 | Natural pelvic movement amplitude |
| MAXDIFF | Maximum pelvic height difference |
| MINDIFF | Minimum pelvic height difference |
Footnotes
Equithrive, Equithrive, Lexington, Ky.
Resverasyn, Equithrive, Lexington, Ky.
Random sequence generator. Available at: random.org. Accessed Jan 31, 2013.
Lameness locator, Equinosis LLC, Columbia, Mo.
Carbocaine, Zoetis Inc, Kalamazoo, Mich.
Kenalog, Bristol-Myers Squibb, New York, NY.
Equi-bute injectable, Aspen Veterinary Resources Ltd, Liberty, Mo.
SAS, version 9.4, SAS Institute Inc, Cary, NC.
Epitools epidemiological calculators. AusVet Animal Health Services and Australian Biosecurity Cooperative Research Centre for Emerging Infectious Disease. Available at: epitools.ausvet.com.au. Accessed Dec 15, 2015.
References
1. Rossdale PD, Hopes R, Digby NJ, et al. Epidemiological study of wastage among racehorses 1982 and 1983. Vet Rec 1985;116:66–69.
2. Pearson W, Orth MW, Lindinger MI. Evaluation of inflammatory responses induced via intra-articular injection of interleukin-1 in horses receiving a dietary nutraceutical and assessment of the clinical effects of long-term nutraceutical administration. Am J Vet Res 2009;70:848–861.
3. Caron JP. Intra-articular injections for joint disease in horses. Vet Clin North Am Equine Pract 2005;21:559–573.
4. McIlwraith CW, Vachon A. Review of pathogenesis and treatment of degenerative joint disease. Equine Vet J Suppl 1988;6:3–11.
5. Kramer CM, Tsang AS, Koenig T, et al. Survey of the therapeutic approach and efficacy of pentosan polysulfate for the prevention and treatment of equine osteoarthritis in veterinary practice in Australia. Aust Vet J 2014;92:482–487.
6. Jaeschke G, Steinbach W. Causal treatment of arthrosis deformans in horses with glucosaminsulfate. Dtsch Tierarztl Wochenschr 1982;89:288–293.
7. Neil KM, Caron JP, Orth MW. The role of glucosamine and chondroitin sulfate in treatment for and prevention of osteoarthritis in animals. J Am Vet Med Assoc 2005;226:1079–1088.
8. Ferris DJ, Frisbie DD, McIlwraith CW, et al. Current joint therapy usage in equine practice: a survey of veterinarians 2009. Equine Vet J 2011;43:530–535.
9. Ramey DW. Skeptical of treatment with glucosamine and chondroitin sulfate (lett). J Am Vet Med Assoc 2005;226:1798–1799.
10. Richardson DW, Loinaz R. An evidence-based approach to selected joint therapies in horses. Vet Clin North Am Equine Pract 2007;23:443–460.
11. Trumble TN. The use of nutraceuticals for osteoarthritis in horses. Vet Clin North Am Equine Pract 2005;21:575–597.
12. Holthoff JH, Woodling KA, Doerge DR, et al. Resveratrol, a dietary polyphenolic phytoalexin, is a functional scavenger of peroxynitrite. Biochem Pharmacol 2010;80:1260–1265.
13. Leiro J, Arranz JA, Parama A, et al. In vitro effects of the polyphenols resveratrol, mangiferin and (-)-epigallocatechin-3-gallate on the scuticociliate fish pathogen Philasterides dicentrarchi. Dis Aquat Organ 2004;59:171–174.
14. Opie LH, Lecour S. The red wine hypothesis: from concepts to protective signalling molecules. Eur Heart J 2007;28:1683–1693.
15. de la Lastra CA, Villegas I. Resveratrol as an antioxidant and pro-oxidant agent: mechanisms and clinical implications. Biochem Soc Trans 2007;35:1156–1160.
16. Harikumar KB, Aggarwal BB. Resveratrol: a multitargeted agent for age-associated chronic diseases. Cell Cycle 2008;7:1020–1035.
17. Renaud S, de Lorgeril M. Wine, alcohol, platelets, and the French paradox for coronary heart disease. Lancet 1992;339:1523–1526.
18. Kopp P. Resveratrol, a phytoestrogen found in red wine. A possible explanation for the conundrum of the ‘French paradox’? Eur J Endocrinol 1998;138:619–620.
19. Csaki C, Keshishzadeh N, Fischer K, et al. Regulation of inflammation signalling by resveratrol in human chondrocytes in vitro. Biochem Pharmacol 2008;75:677–687.
20. Shakibaei M, Csaki C, Nebrich S, et al. Resveratrol suppresses interleukin-1beta-induced inflammatory signaling and apoptosis in human articular chondrocytes: potential for use as a novel nutraceutical for the treatment of osteoarthritis. Biochem Pharmacol 2008;76:1426–1439.
21. Elmali N, Baysal O, Harma A, et al. Effects of resveratrol in inflammatory arthritis. Inflammation 2007;30:1–6.
22. Liang Q, Wang XP, Chen TS. Resveratrol protects rabbit articular chondrocyte against sodium nitroprusside-induced apoptosis via scavenging ROS. Apoptosis 2014;19:1354–1363.
23. Jin H, Liang Q, Chen T, et al. Resveratrol protects chondrocytes from apoptosis via altering the ultrastructural and biomechanical properties: an AFM study. PLoS ONE 2014;9:e91611.
24. Liu L, Gu H, Liu H, et al. Protective effect of resveratrol against IL-1beta-induced inflammatory response on human osteoarthritic chondrocytes partly via the TLR4/MyD88/NF-kappaB signaling pathway: an “in vitro study.” Int J Mol Sci 2014;15:6925–6940.
25. Kim HJ, Braun HJ, Dragoo JL. The effect of resveratrol on normal and osteoarthritic chondrocyte metabolism. Bone Joint Res 2014;3:51–59.
26. Coradini K, Friedrich RB, Fonseca FN, et al. A novel approach to arthritis treatment based on resveratrol and curcumin coencapsulated in lipid-core nanocapsules: in vivo studies. Eur J Pharm Sci 2015;78:163–170.
27. Li W, Cai L, Zhang Y, et al. Intra-articular resveratrol injection prevents osteoarthritis progression in a mouse model by activating SIRT1 and thereby silencing HIF-2alpha. J Orthop Res 2015;33:1061–1070.
28. Guide for veterinary services and judging of equestrian events. Lexington, Ky: American Association of Equine Practitioners, 1991;19.
29. Keegan KG, Wilson DA, Wilson DJ, et al. Evaluation of mild lameness in horses trotting on a treadmill by clinicians and interns or residents and correlation of their assessments with kinematic gait analysis. Am J Vet Res 1998;59:1370–1377.
30. Keegan KG, Pai PF, Wilson DA, et al. Signal decomposition method of evaluating head movement to measure induced forelimb lameness in horses trotting on a treadmill. Equine Vet J 2001;33:446–451.
31. Labens R, Mellor DJ, Voute LC. Retrospective study of the effect of intra-articular treatment of osteoarthritis of the distal tarsal joints in 51 horses. Vet Rec 2007;161:611–616.
32. de Grauw JC, Visser-Meijer MC, Lashley F, et al. Intra-articular treatment with triamcinolone compared with triamcinolone with hyaluronate: a randomised open-label multicentre clinical trial in 80 lame horses. Equine Vet J 2016;48:152–158.
33. Friedman LM. Blindness. In: Fundamentals of clinical trials. 4th ed. New York: Springer, 2010;119.
34. Schulz KF, Chalmers I, Hayes RJ, et al. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995;273:408–412.
35. Dabareiner RM, Carter GK, Dyson SJ. The tarsus. In: Ross MW, Dyson SJ, eds. Diagnosis and management of lameness in the horse. Philadelphia: Saunders, 2003; 440–449.
36. Keegan KG, Kramer J, Yonezawa Y, et al. Assessment of repeatability of a wireless, inertial sensor-based lameness evaluation system for horses. Am J Vet Res 2011;72:1156–1163.
37. Keegan KG, MacAllister CG, Wilson DA, et al. Comparison of an inertial sensor system with a stationary force plate for evaluation of horses with bilateral forelimb lameness. Am J Vet Res 2012;73:368–374.
38. Donnell JR, Frisbie DD, King MR, et al. Comparison of subjective lameness evaluation, force platforms and an inertialsensor system to identify mild lameness in an equine osteoarthritis model. Vet J 2015;206:136–142.
39. Keegan KG, Wilson DA, Kramer J, et al. Comparison of a body-mounted inertial sensor system-based method with subjective evaluation for detection of lameness in horses. Am J Vet Res 2013;74:17–24.
40. McCracken MJ, Kramer J, Keegan KG, et al. Comparison of an inertial sensor system of lameness quantification with subjective lameness evaluation. Equine Vet J 2012;44:652–656.
41. Kramer J, Keegan KG, Kelmer G, et al. Objective determination of pelvic movement during hind limb lameness by use of a signal decomposition method and pelvic height differences. Am J Vet Res 2004;65:741–747.
42. Du J, White N, Eddington ND. The bioavailability and pharmacokinetics of glucosamine hydrochloride and chondroitin sulfate after oral and intravenous single dose administration in the horse. Biopharm Drug Dispos 2004;25:109–116.
43. Soo E, Thakur S, Qu Z, et al. Enhancing delivery and cytotoxicity of resveratrol through a dual nanoencapsulation approach. J Colloid Interface Sci 2015;462:368–374.
44. Ememe MU, Mshelia WP, Ayo JO. Ameliorative effects of resveratrol on oxidative stress biomarkers in horses. J Equine Vet Sci 2015;35:518–523.
45. McIlwraith CW. The use of intra-articular corticosteroids in the horse: what is known on a scientific basis? Equine Vet J 2010;42:563–571.
46. Kerbyson N. Recent initiatives in evidence-based veterinary medicine. Equine Vet J 2015;47:503–504.
47. Sandersen C, Olejnik D, Franck T, et al. Inhalation with NDS27 attenuates pulmonary neutrophilic inflammation in recurrent airway obstruction. Vet Rec 2011;169:100.
48. Derochette S, Franck T, Mouithys-Mickalad A, et al. Curcumin and resveratrol act by different ways on NADPH oxidase activity and reactive oxygen species produced by equine neutrophils. Chem Biol Interact 2013;206:186–193.
49. Kohnen S, Franck T, Van Antwerpen P, et al. Resveratrol inhibits the activity of equine neutrophil myeloperoxidase by a direct interaction with the enzyme. J Agric Food Chem 2007;55:8080–8087.
50. Latruffe N, Lancon A, Frazzi R, et al. Exploring new ways of regulation by resveratrol involving miRNAs, with emphasis on inflammation. Ann N Y Acad Sci 2015;1348:97–106.
51. Wendling D, Abbas W, Godfrin-Valnet M, et al. Resveratrol, a sirtuin 1 activator, increases IL-6 production by peripheral blood mononuclear cells of patients with knee osteoarthritis. Clin Epigenetics 2013;5:10.
52. Holmes MA, Ramey DW. An introduction to evidence-based veterinary medicine. Vet Clin North Am Equine Pract 2007;23:191–200.
