The biomechanical role of SF in horses, which are considered to have a high-performance locomotor system, is important to understand from both an equine and human medical perspective.1,2 This viscous ultrafiltrate of plasma is in direct contact with articular cartilage, providing lubrication of cartilage surfaces and protection of articular joints.3,4 In addition, SF serves as a biochemical pool through which nutrients and regulatory cytokines traverse.
Synovial fluid contains molecules that provide low-friction, low-wear properties to articulating cartilage surfaces.5 Among molecules postulated to play a key role in lubrication, alone or in combination with others, is HA.6 Hyaluronidase is a glycosaminoglycan synthesized by chondrocytes and synovial fibroblasts.7 In joints, HA is a major component of SF and the extracellular matrix of cartilage.8 Among other functions, this molecule serves as a lubricant and plays a major role in shock absorption and distribution of mechanical forces impacting a joint. Moreover, HA inhibits the migration of inflammatory cells9 as well as cytokine-induced production of matrix metalloproteinases and other proinflammatory mediators10; it may also provide analgesic effects through interaction with intra-articular pain receptors.11
Substantial advances in the veterinary field have provided scientists with novel ways to examine macromolecular interactions in SF and other complex biological fluids. The rheological properties (ie, viscoelastic properties) of equine SF were first evaluated in the 1970s.12 However, the first study13 to characterize the viscosity of SF in orthopedically normal horses (ie, normal SF) and the role of protein concentration was reported in the 1990s.13 In general, normal SF is a non-Newtonian fluid, with high viscoelasticity and strong shear-thinning properties. Although several components of human SF are known to have a substantial impact on the lubricating properties of articular joints,14 limited information is available regarding the rheological properties of equine SF.15 The purposes of the study reported here were to investigate the complex nature of SF in orthopedically normal horses and to determine whether viscoelastic properties of this fluid would differ between Thoroughbreds and warmblood horses.
Materials and Methods
Animals
Forty-five warmblood horses and 30 Thoroughbreds with a median age of 9.5 years (range, 4 to 16 years) were included in the study and considered as 2 breed groups. All included warmblood horses and Thoroughbreds originated from the same premises (horse riding school and Athens racetrack, respectively) and underwent the same training protocol. Mean ± SD body weight was 495 ± 65 kg for Thoroughbreds and 570 ± 60 kg for warmblood horses, and mean ± SD height at withers (highest point of the shoulders) was 165 ± 8 cm and 162 ± 10 cm, respectively.
All horses received a thorough clinical examination and a hematologic analysis, including blood cytologic evaluation, total WBC count, and Hct measurement. All horses were confirmed to be free of signs of lameness at a walk and trot and had negative results of joint flexion testing. All forelimb joints from which SF samples were to be collected were randomly selected, radiographed, and confirmed free of radiographic abnormalities by consensus opinion of 2 experienced veterinary radiologists (1 of whom was board certified), who interpreted the radiographs without knowledge of the horse's health status. The study protocol used in this project complied with the guidelines for the use of animals in research. Horses were included if the owner provided written consent and was deemed to be willing and capable of complying with the requirements of the study. None of the horses had received intra-articular medications or NSAIDs during the 3 months prior to SF aspiration.
Sample collection
Synovial fluid samples were sterilely collected from 1 metacarpophalangeal (fetlock) and intercarpal joint (limb arbitrarily selected) of all horses via 21-gauge needle into a 2.5-mL syringe, as reported elsewhere.11 Samples were also collected from the distal interphalangeal (coffin) joint of 15 warmblood horses. The intercarpal joint was entered via a dorsal approach, with the limb held and the carpus flexed. The joint capsule was penetrated at a depth of approximately 1.3 cm. Arthrocentesis of the fetlock joint was performed with the limb in flexion and the needle inserted into the lateral aspect of the palmar pouch. The joint capsule was penetrated at a depth of approximately 1 cm. The coffin joint was entered via a dorsal approach, with the limb bearing weight. The needle was inserted into the dorsal pouch of the joint at the point where the pastern joint met the proximal edge of the coronet. Collected samples were transferred to tubes containing EDTA for routine SF analysis (total protein concentration and cytologic evaluation) and to 1.0-mL collection tubes for rheological evaluation within 6 hours after aspiration.
SF analysis
The viscosity of SF samples was evaluated subjectively as described elsewhere.16 Full rheological characterization was also performed by use of a controlled-stress rheometer,a set at a 40-mm parallel plate geometry, frequency of 0.5 Hz, and strain of 3%. The frequency of 0.5 Hz was selected because, in the authors' experience, higher frequencies may result in data artifacts, whereas lower frequencies result in sample drying. The properties of the SF samples were evaluated at a temperature of 37.5°C (to mimic a horse's body temperature), with an accuracy of 0.1°C provided by the Peltier plate temperature control system of the rheometer. Dynamic oscillatory experiments, in which a shearing sinusoidal deformation was applied to the SF samples in the linear viscoelastic range, were performed to measure the storage modulus G' and loss modulus G“. The G' describes the elastic character of the fluid, reflecting quantitatively the energy that is stored in the material during the deformation process. On the other hand, G” represents the viscous property of the material, and its value corresponds to the energy, in particular heat, that is dissipated from the material during deformation.
Synovial fluid samples were prepared for microscopic evaluation by use of the cytocentrifuge function of an automated slide stainer.b All prepared slides were stained with Giemsa stain and then examined by 2 experienced clinical veterinary pathologists for nucleated cell classification, cell morphological evaluation, and microorganism detection. A minimum of 300 nucleated cells was examined, and the percentages of neutrophils, mononuclear cells, and unrecognizable cells were recorded for the differential NCC. An automated analyzerc was used to measure the total NCC in each sample.
Refractometryd was performed to measure total solids concentration in SF samples as an estimate of total protein concentration. Quantification of HA concentration in 30 randomly selected samples obtained from each breed group was also performed with a commercial ELISA kit.e Samples were first diluted 1:1,000 in sample diluent, and then the assay was performed in accordance with the manufacturer's instructions. Absorbance values were recorded by use of a microplate photometer.f Analysis of the absorbance values of SF samples and calibrators was performed with the aid of a commercial software package.g Synovial fluid samples with HA concentration exceeding the dynamic range of quantification of the method were reanalyzed at a dilution 1:2,000.
Statistical analysis
All statistical analyses were conducted with the aid of statistical software,h and all tests were 2 tailed. To evaluate the effect of factors (breed group and joint type) on the mean values of the response variables (G' and G“), linear mixed modeling was performed.i The optimal fixed component structure of this model was defined through a backward elimination procedure,j by which the highest-order interactions were tested first and, if a significant association with the response variable was identified, the lower-order effects were not tested. The P values for the fixed component of the model were calculated from an F test on the basis of the Kenward-Roger approximation of the degrees of freedom.17 Finally, graphical validation was used to assess the underlying assumptions of homoscedasticity and normality of residuals of the selected models. For all tests, a difference was considered as significant at P < 0.05.
Results
G'
Linear mixed modeling revealed a significant main effect of joint type (P < 0.001) but not breed group (P = 0.34) on logarithmically transformed G' values. A significant interaction effect on these values was also identified between breed group and joint type (P = 0.047; Figure 1; Supplementary Table SI, available at avmajournals.avma.org/doi/suppl/10.2460/ajvr.80.4.342). Post hoc analysis revealed significant differences between Thoroughbreds and warmblood horses for the fetlock joint (P = 0.006), between the fetlock and intercarpal joints of Thoroughbreds (P = 0.04), and between the fetlock and coffin joints (P < 0.001) and intercarpal and coffin joints (P = 0.02) of warmblood horses.
Mean logarithmically transformed G' (elastic modulus) values for SF samples obtained from the metacarpophalangeal (fetlock), intercarpal (carpus), and distal interphalangeal (coffin) joints of 30 Thoroughbreds (TB) and 45 warmblood horses (WB). Mixed linear effects modeling revealed a significant effect on these values of joint type (P < 0.001) and the interaction between breed and joint type (P = 0.047) but not of breed (P = 0.34). Whiskers represent 95% confidence intervals.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.342
Mean logarithmically transformed G' (elastic modulus) values for SF samples obtained from the metacarpophalangeal (fetlock), intercarpal (carpus), and distal interphalangeal (coffin) joints of 30 Thoroughbreds (TB) and 45 warmblood horses (WB). Mixed linear effects modeling revealed a significant effect on these values of joint type (P < 0.001) and the interaction between breed and joint type (P = 0.047) but not of breed (P = 0.34). Whiskers represent 95% confidence intervals.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.342
Mean logarithmically transformed G' (elastic modulus) values for SF samples obtained from the metacarpophalangeal (fetlock), intercarpal (carpus), and distal interphalangeal (coffin) joints of 30 Thoroughbreds (TB) and 45 warmblood horses (WB). Mixed linear effects modeling revealed a significant effect on these values of joint type (P < 0.001) and the interaction between breed and joint type (P = 0.047) but not of breed (P = 0.34). Whiskers represent 95% confidence intervals.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.342
G“
Linear mixed modeling revealed a significant main effect of joint type (P < 0.001) but not breed (P = 0.52) on logarithmically transformed G' values. A significant interaction effect on these values was also identified between breed and joint type (P = 0.04; Figure 2; Supplementary Table S2, available at avmajournals.avma.org/doi/suppl/10.2460/ajvr.80.4.342). Post hoc analysis revealed significant differences between Thoroughbreds and warmblood horses for the fetlock joint (P = 0.01), between the fetlock and intercarpal joints of Thoroughbreds (P = 0.02), and between the fetlock and coffin joints (P < 0.001) and intercarpal and coffin joints (P = 0.02) of warmblood horses.
Mean logarithmically transformed G“ (viscous modulus) values for SF samples from the joints and horses of Figure 1. Mixed linear effects modeling revealed a significant effect on these values of joint type (P < 0.001) and the interaction between breed and joint type (P = 0.04) but not of breed (P = 0.47). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.342
Mean logarithmically transformed G“ (viscous modulus) values for SF samples from the joints and horses of Figure 1. Mixed linear effects modeling revealed a significant effect on these values of joint type (P < 0.001) and the interaction between breed and joint type (P = 0.04) but not of breed (P = 0.47). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.342
Mean logarithmically transformed G“ (viscous modulus) values for SF samples from the joints and horses of Figure 1. Mixed linear effects modeling revealed a significant effect on these values of joint type (P < 0.001) and the interaction between breed and joint type (P = 0.04) but not of breed (P = 0.47). See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 80, 4; 10.2460/ajvr.80.4.342
Biochemical analysis and NCC
Total NCC, total protein concentration, and HA concentrations in SF samples from Thoroughbred and warmblood horses were summarized (Table 1). No significant differences in analyte values were identified within or between breed groups.
Summary data for measurements of HA and total protein concentration and NCC in SF samples obtained from the metacarpophalangeal (fetlock), intercarpal, and distal interphalangeal (coffin) joints of orthopedically normal Thoroughbreds (n = 15) and warmblood horses (15).
| Measurement | Breed group | Fetlock joint | Intercarpal joint | Coffin joint |
|---|---|---|---|---|
| Sample volume (mL) | Thoroughbred | 8 ± 1 | 7 ± 1 | — |
| Warmblood | 8 ± 2 | 6 ± 1 | 3 ± 1 | |
| HA (μg/mL) | Thoroughbred | 839 (613–1,230) | 702 (648–966) | — |
| Warmblood | 931 (668–1,479) | 877 (552–1,021) | 571 (317–939) | |
| TP (g/dL) | Thoroughbred | 1.4 ± 0.2 | 1.3 ± 0.2 | — |
| Warmblood | 1.6 ± 0.3 | 1.5 ± 0.3 | 1.2 ± 0.2 | |
| NCC (cells/μL) | Thoroughbred | 129 ± 61 | 101 ± 35 | — |
| Warmblood | 103 ± 50 | 90 ± 46 | 114 ± 36 |
Values for HA are median (range) and for all other measurements are mean ± SD. - = Not measured. TP = Total protein.
Discussion
In the present study, several differences in the viscoelastic properties (G' and G“) of SF were identified between orthopedically normal warmblood horses and orthopedically normal Thoroughbreds and among fetlock, intercarpal, and coffin joints within the same breed group. In sport horses, lameness is approximately 3 times as common in the forelimbs as in the hind limbs, and 95% of forelimb lameness occurs within or distal to the carpus.18 For this reason, we chose to focus on forelimb joints for our study.
In contrast with other large animals such as cattle, horses are dependent on healthy functioning of their complex musculoskeletal system to perform optimally, maintaining the minimum energy cost during locomotion at high loads. The thin nature of equine transmitted by the fetlock joint of horses are higher than those acting at any other joint in the distal portion of the forelimb.22 In the study reported here, the higher G' and G“ values for SF fluid samples obtained from the fetlock joint, compared with those for the coffin joint, could have been attributable to the need for SF with stronger viscoelastic properties for that joint to be able to absorb forces it sustains.
Synovial fluid samples from the coffin joint of warmblood horses in the present study had the lowest G' and G“ values of all joints evaluated. The smallest of the 3 joints evaluated, the coffin joint is encased inside the hoof and is flexed by the strong deep digital flexor tendon attached to the back of the coffin bone. The navicular bone acts as a fulcrum over which the deep digital flexor tendon runs. Joint power analysis has shown that the coffin joint serves an energy-absorbing function23; hence, the observed low viscoelastic values could have been attributable to the anatomic and functional role of the coffin joint.
The G' and G“ values for SF samples from the fetlock joint were significantly lower for Thoroughbreds than for warmblood horses in the present study. However, given the small number of evaluated horses and the lack of significant differences between breed groups for the other evaluated joint, cautious interpretation of this finding is recommended.
The purpose of the other assessments of SF properties (total protein and HA concentration measurement and NCC) in the study reported here was to detect any possible biochemical and cytologic differences between breed groups or among joints. A wide range of values was observed, presumably owing to individual variation and age differences of included horses, and no significant differences were detected.
The limited number of horses included in the present study should be considered when other investigators attempt to extrapolate these findings to the general population of orthopedically normal, active athletic horses. However, to the authors' knowledge, the present study represented the first of its kind, and we believe that the information obtained in this preliminary exploration could be valuable and form the basis for future studies.
Acknowledgments
This manuscript represents a portion of a larger project conducted by Panagiota Tyrnenopoulou at the Aristotle University of Thessaloniki as partial fulfillment of the requirements for a Doctorate of Philosophy.
The authors declare that there were no conflicts of interest.
The authors acknowledge Serafeim Chaintoutis for performance of ELISA measurements.
ABBREVIATIONS
| HA | Hyaluronic acid |
| NCC | Nucleated cell count |
| SF | Synovial fluid |
Footnotes
AR-G2, TA Instruments, Crawley, England.
Aerospray Hematology Pro slide stainer, Wescor Biomedical Systems, Logan, Utah.
Scil Vet ABC animal blood counter, Scil Animal Care Company, Holtzheim, France.
Model T2-NE-Clinical, Atago Ltd, Tokyo, Japan.
Teco HA plus, Tecomedical AG, Sissach, Switzerland.
Stat Fax 3200, Awareness Technology Inc, Palm City, Fla.
GraphPad Prism, version 7.03, GraphPad Software Inc, La Jolla, Calif.
R, version 3.1.2, R Foundation for Statistical Computing, Vienna, Austria.
lme4: Linear Mixed-Effects Models using ‘Eigen’ and S4, version 1.1–12, Bates D, Maechler M, Bolker B, et al. Available at: CRAN.R-project.org/package=lme4. Accessed April 16, 2016.
lmerTest: Tests in Linear Mixed Effects Models, R package version 1.1–7. Kuznetsova A, Brockhoff PB, Christensen RHB. Available at: CRAN.R-project.org/ackage=lmerTest. Accessed Mar 18, 2014.
References
1. Chu CR, Szczodry M, Bruno S. Animal models for cartilage regeneration and repair. Tissue Eng Part B Rev 2010; 16:105–115.
2. McIlwraith CW, Frisbie DD, Kawcak CE. The horse as a model of naturally occurring osteoarthritis. Bone Joint Res 2012;1:297–309.
3. Hui AY, McCarty WJ, Masuda K, et al. A systems biology approach to synovial joint lubrication in health, injury, and disease. Wiley Interdiscip Rev Syst Biol Med 2012;4:15–37.
4. Johnston JP. The viscosity of normal and pathological human synovial fluids. Biochem J 1955;59:633–637.
5. Blewis ME, Nugent-Derfus GE, Schmidt TA, et al. A model of synovial fluid lubricant composition in normal and injured joints. Eur Cell Mater 2007;13:26–39.
6. Mazzucco D, Scott R, Spector M. Composition of joint fluid in patients undergoing total knee replacement and revision arthroplasty: correlation with flow properties. Biomaterials 2004;25:4433–4445.
7. Iwanaga T, Shikichi M, Kitamura H, et al. Morphology and functional roles of synoviocytes in the joint. Arch Histol Cytol 2000;63:17–31.
8. Fraser JR, Laurent TC, Laurent UB. Hyaluronan: its nature, distribution, functions and turnover. J Intern Med 1997;242:27–33.
9. Forrester JV, Wilkinson PC. Inhibition of leukocyte locomotion by hyaluronic acid. J Cell Sci 1981;48:315–331.
10. Wang CT, Lin YT, Chiang BL, et al. High molecular weight hyaluronic acid down-regulates the gene expression of osteoarthritis-associated cytokines and enzymes in fibroblast-like synoviocytes from patients with early osteoarthritis. Osteoarthritis Cartilage 2006;14:1237–1247.
11. Gomis A, Miralles A, Schmidt RF, et al. Nociceptive nerve activity in an experimental model of knee joint osteoarthritis of the guinea pig: effect of intra-articular hyaluronan application. Pain 2007;130:126–136.
12. Rejnö S. Viscosity of equine synovial fluid. Acta Vet Scand 1976;17:169–177.
13. Korenek NL, Andrews FM, Maddux JM, et al. Determination of total protein concentration and viscosity of synovial fluid from the tibiotarsal joints of horses. Am J Vet Res 1992;53:781–784.
14. Crockett R. Boundary lubrication in natural articular joints. Tribol Lett 2009;35:77–84.
15. Borzacchiello A, Mayol L, Schiavinato A, et al. Effect of hyaluronic acid amide derivative on equine synovial fluid viscoelasticity. J Biomed Mater Res A 2010;92:1162–1170.
16. Steel CM. Equine synovial fluid analysis. Vet Clin North Am Equine Pract 2008;24:437–454.
17. Kenward MG, Roger JH. Small sample inference for fixed effects from restricted maximum likelihood. Biometrics 1997;53:983–997.
18. Clayton HM, Chateau H, Back W. Forelimb function. In: Equine locomotion. London: Saunders Elsevier, 2013;99–125.
19. Dimery NJ, Alexander RMN, Ker RF. Elastic extension of leg tendons in the locomotion of horses (Equus caballus). J Zool 1986;210:415–425.
20. Schauder W. Die besonderen stoßbrechenden Einrichtungen an den Gliedmaßen des Pferdes. Dtsch Tierarztl Wochenschr 1952;59:35–38.
21. Back W, Schamhardt HC, Barneveld A. Are kinematics of the walk related to the locomotion of a warmblood horse at the trot? Vet Q 1996;18:79–84.
22. Harrison SM, Whitton RC, Kawcak CE, et al. Relationship between muscle forces, joint loading and utilization of elastic strain energy in equine locomotion. J Exp Biol 2010;213:3998–4009.
23. Clayton HM, Lanovaz JL, Schamhardt HC, et al. Net joint moments and powers in the equine forelimb during the stance phase of the trot. Equine Vet J 1998;30:384–389.
