Acellular Urinary Bladder Matrix in a Collagenase Model of Superficial Digital Flexor Tendonitis in Horses

ORIGINAL RESEARCH Acellular Urinary Bladder Matrix in a Collagenase Model of Superficial Digital Flexor Tendonitis in Horses Ty W. Wallis, DVM, MS, DACVS,a Gary M. Baxter, VMD, MS, DACVS,b Natasha M. Werpy, DVM, DACVR,c,d Gary L. Mason, DVM, PhD, DACVP,e David D. Frisbie, DVM, PhD, DACVS,d and Nicolai Jarloev, DVM, PhDf

ABSTRACT The objective of the present study was to determine the efficacy of urinary bladder matrix (UBM) in collagenaseinduced superficial digital flexor (SDF) tendonitis by using clinical, ultrasonographic, and histologic data. A total of eight healthy adult horses were used in this study. Bilateral forelimb SDF tendonitis was created in the horses by injecting collagenase. After 14 days, one randomly selected forelimb SDF tendon was blindly treated with UBM and the opposite tendon was treated with a control (saline). Clinical and ultrasonographic parameters including lameness, lesion size, ultrasonographic fiber pattern, and echogenicity were measured throughout the study. After 84 days, horses were euthanized and SDF tendon lesions from the two groups were compared statistically using an analysis of variance with significance set at P % .05.Results showed that there were no significant differences between the treated and control tendons for any of the clinical, ultrasonographic, gross, or histologic variables. UBM does not appear to be an effective treatment for collagenaseinduced SDF tendonitis. However, there may be differences in clinical tendonitis that might render the treatment more effective in the clinical setting. Keywords: Urinary bladder matrix; Tendonitis; Horse; Collagenase From the Pinnacle Equine Hospital, Louisville, TNa; Department of Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, COb; Department of Environmental Health and Radiological Sciences, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, COc; Equine Orthopaedic Research Center, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, COd; Department of Microbiology, Immunology, and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, COe; and Hoersholm Horseclinic, Fredensborg, Denmarkf. Results of this project were presented at the 53rd Annual Convention of the American Association of Equine Practitioners in Orlando, Florida on December 3, 2007. Reprint requests: Ty W. Wallis, DVM, MS, DACVS, Pinnacle Equine Hospital, 1958 Jones Bend Rd., Louisville, TN 37777. 0737-0806/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.jevs.2010.05.009

Journal of Equine Veterinary Science  Vol 30, No 7 (2010)

INTRODUCTION Tendonitis may result from a severe strain to the superficial digital flexor (SDF) tendon, because of excessive loading and overstretching of the tendon, repetitive overuse, or microdamage of the tendon associated with exercise.1-5 Strain-induced injuries are believed to result following a phase of molecular degeneration or an inflammatory event that is neither clinically evident nor produces any reparative responses but instead progressively weakens the structure.5 Tendonopathies often begin with degeneration of and minor changes in the structural integrity of the tendon, and thus predispose an already high-risk structure to injury. After the structural strength is overcome, physical disruption occurs within the tendon matrix. Various structural breakdowns are known to occur including fibrillar stretching with breakage of crosslinks, fibrillar rupture, or in some severe cases, separation of tendon tissue.3 The tendon on healing produces a collagenous scar, predominantly of type III collagen, which is generally proportional to the severity of the tendon lesion.6 Tendonitis of the SDF tendon is a common cause of lameness in performance horses, especially race horses and event horses. Risk factors that contribute toward injury are the speed of exercise, the small cross-sectional area (CSA) of the SDF tendon, and the excessive load that is placed repetitively on the tendon during the early and mid-stance phase of the stride.3,6 Additionally, predisposing factors for SDF tendonitis in other types of performance horses include inadequate training, muscle fatigue, uneven and slippery ground, sudden turning, excessive pastern slope, improper shoeing, and the long toe–low heel hoof conformation.6 Appropriate treatment of tendonitis after resolving the initial inflammatory stage is controversial. Treatment options that have been used for tendonitis include administration of intralesional hyaluronan and b-aminoproprionitrile fumurate, and intramuscular polysulfated glycosaminoglycans; tendon splitting with or without superior check ligament desmotomy; use of non-steroidal anti-inflammatory drugs; controlled exercise alone; and various physical therapy modalities such as ultrasound, laser, and magnetic therapies.1,2,6-11 More recent treatment options include

365

366

extracorporeal shockwave and intralesional bone marrow therapy, insulin-like growth factor 1 hormone therapy, autogenous mesenchymal stem cells therapy, and administration of urinary bladder matrix (UBM) powder (ACell Vet Powder, ACell, Inc., Columbia, MD).2,7,8,12,13 UBM is a lyophilized powder derived from the extracellular matrix of the basement membrane of swine urinary bladder.14-16 It is acellular in structure but is thought to recruit regenerative cells and other necessary growth factors from the circulatory system and local tissues for the purpose of tissue differentiation.12,14-18 In other species, UBM has been found to produce a profound angiogenic response in the first 5 to 7 days post treatment.15,17 The potential benefits of UBM to tendon healing include providing a scaffold for collagen deposition within the damaged tendon, recruiting growth factors to the site of injury, and minimizing excessive fibrous tissue formation.12,14,16,18 Although little is known about the immunogenicity of UBM, a cross-species immune response may be possible, as this product originates from swine bladder.18 However, to our knowledge, this has not yet been documented clinically. Definitive information on the benefits of intralesional UBM therapy in horse tendon injuries is scarce. Anecdotally, it has shown promise with both tendonitis and desmitis in clinical cases. In a recent report of 53 horses treated with intralesional UBM, 81% of the horses R6 months post-treatment were found be sound, healthy, and capable of carrying out work.12 As compared with more conventional treatments, tendon and ligament healing was thought to occur more rapidly with better quality of the repaired tissue being visible ultrasonographically. Less scarring and permanent enlargement of the treated tendons and ligaments were thought to occur clinically.12 However, no controlled studies evaluating the efficacy of intralesional UBM in the treatment of soft-tissue injuries in horses have been performed. The purpose of our study was to determine the efficacy of intralesional UBM in horses with collagenase-induced tendonitis, using clinical, ultrasonographic, and histologic data. We hypothesized that as compared with the saline control, UBM would promote better quality tissue repair in the collagenase-induced tendonitis model.

MATERIALS AND METHODS The study protocol was approved by the Colorado State University Institutional Animal Care and Use Committee. For this study, eight healthy adult horses free from lameness and musculoskeletal abnormalities were selected on the basis of baseline physical and lameness examinations. The cohort consisted of 5 females, 2 stallions, and 1 gelding, with a mean age of 4.25 years (range, 2.5–5) and a mean body weight of 395 kg (range, 350–425 kg).

TW Wallis et al  Vol 30, No 7 (2010)

Baseline ultrasound was performed on the front limbs, and CSA was measured in five zones (6 cm, 8 cm, 9 cm, 10 cm, and 12 cm distal to the palmar aspect of the carpometacarpal joint as assessed with ultrasound, respectively). Baseline limb circumference measurements were taken at 9 cm distal to the palmar aspect of the carpometacarpal joint as assessed with ultrasound, using a flexible tape measure, and calipers were used to measure the medial-lateral width of the SDF tendon at the same level. On day 14, after pretreating with flunixin meglumine (1.1 mg/kg IV) and performing a high 4-point nerve block with bupivacaine, tendonitis was induced in the front limbs. Collagenase (Bacterial Collagenase Type I, Sigma-Aldrich Co, St. Louis, MO; 2,100 units) diluted in 0.3 mL sterile water was injected using a 27 gauge 1/200 needle into the body of the SDF tendons, under ultrasonographic guidance at two sites (ie, 8 cm and 10 cm) distal to the carpometacarpal joint, as has been described in previous studies.9,13,19 Injections were administered at the medial aspect of the tendon so as to visualize the entrance of the needle into the thicker portion of the tendon and its advancement to the center of the tendon under ultrasonographic guidance from a palmar transverse view. After 12 hours, a high 4-point block with bupivacaine was repeated, and each horse was treated with phenylbutazone (4.4 mg/kg IV q 12 h for 2 doses, then 2.2 mg/kg PO q 12 h for 5 days). The limbs were then hosed with cold water for 15 minutes once daily for 3 days and bandaged until treatment to minimize swelling and inflammation. On the basis of pilot studies, horses were housed in stalls for the initial 2 weeks after collagenase injection to attempt to allow for lesion stabilization. After 14 days (day 0), lameness examinations were performed by using the American Association of Equine Practitioners lameness scale.20 The severity of pain on palpation of the SDF tendon was determined by using a grading scale ranging from 0 to 4 (0 ¼ no pain, 1 ¼ slight pain, 2 ¼ mild pain, 3 ¼ moderate pain, and 4 ¼ marked pain). Front limb circumference and SDF tendon width at the site of collagenase injections were measured. Ultrasound evaluations were performed in both the transverse and longitudinal planes to document the size of the hypoechoic lesion and severity of SDF tendonitis. The CSA of the lesion calculated as a percentage of CSA of the tendon was determined in the transverse plane, and the length of lesion was determined in the longitudinal plane. Ultrasound scores were recorded separately for fiber alignment and echogenicity on a scale ranging from 0 to 4 (0 ¼ normal, 1 ¼ slight disruption/hypoechogenicity, 2 ¼ mild disruption/hypoechogenicity, 3 ¼ moderate disruption/hypoechogenicity, and 4 ¼ severe disruption/hypoechogenicity).21 On day 0, one SDF tendon from each horse was randomly assigned to the treatment group (UBM) and the opposite SDF tendon to a sham treatment (saline). The investigators were blinded to the treatments. Using a 22

TW Wallis et al  Vol 30, No 7 (2010)

gauge 100 needle, the treated limb was injected with 4 mL of a suspension created by diluting 0.2 grams of UBM in 6 mL of 0.9% sterile saline.12 A volume of 4 mL was chosen on the basis of the projected average lesion size created with this dose of collagenase from other studies (Dahlgren LA, Virginia-Maryland Regional College of Veterinary Medicine, Blacksburg, VA: Personal communication, 2006). The suspension was distributed throughout the lesion using ultrasonographic guidance in 3 areas as per the manufacturer’s recommendations. Similarly, the control limb was injected with 4 mL of 0.9% saline. The aforementioned protocol of flunixin meglumine, phenylbutazone, bandaging, and cold hosing used after induction of the model was followed again after treatment. For the initial 14 days, the horses were confined to their stalls; they were then hand-walked, 15 minutes once daily, for the next 14 days; and 30 minutes once daily for another 28 days. At 56 days following treatment, the horses were taken to a small paddock for 6 hours daily for the remainder of the 84-day treatment period. Clinical and lameness examinations, as described previously, were documented at 7-day intervals for the initial 28 days, and then followed-up at every 14-day interval. Ultrasound evaluations as aforementioned were obtained at days 14, 28, 56, and 84 post-treatment by an ultrasonographer blinded to the treatment groups. After 84 days posttreatment, the horses were humanely euthanized with an overdose of barbiturate, and the SDF tendons were evaluated grossly and histologically. A three-dimensional measurement of the area of the affected tendon was taken, and the tendons were split longitudinally. The extent of hemorrhage within the tendons on the cut sections was recorded. Sagittal (longitudinal) and transverse histologic samples were collected and examined from four representative areas of the tendon in the area of the original lesion (6 cm, 8 cm, 10 cm, and 12 cm distal to the carpometacarpal joint). These samples were then fixed in formalin, embedded in paraffin, sectioned, and stained with hematoxylin and eosin. They were graded on a scale ranging from 1 to 4 (1 ¼ normal healed tendon, 4 ¼ severely diseased tendon for each category) by a blinded pathologist using the following nine specific parameters: tendon cell shape, tendon cell density, free hemorrhage, neovascularization, inflammatory cell infiltrate, collagen fiber linearity, collagen fiber uniformity, collagen fiber crimping visualized with polarized light microscopy, and epitenon thickening. Statistical Methods Statistical analysis was performed using a commercially available software program (The GLIMMIX procedure; SAS Institute, Cary, NC, 2006) to perform a mixed model analysis of variance, with the treatment group nested within horse as the random variable. Significance was set at P % .05.

367

RESULTS There were no significant differences between any clinical parameters measured in the treated and control limbs. However, in the case of measurement of lameness, there was a trend toward a significant difference between the treated (mean score ¼ 1.21, range 1–4) and control limbs (mean score ¼ 1.44, range 1–4) (P ¼ .1), with the control limbs having higher lameness scores (Table 1). When the treated tendons were compared grossly and histologically with the control tendons, there were no significant differences found between the two groups for any of the response variables evaluated, even when divided into transitional and maximal injury zones. In the case of the treated tendons, lesions tended to be smaller on gross measurements. However, as compared with the control tendons, scores for all histologic variables except epitenon thickness were greater; that is, they were more abnormal. In contrast, there was a trend toward significantly lower (more normal) collagen crimping scores in the case of the control tendons (P ¼ .07) (Table 1). In the case of the CSA of lesions, there was a trend toward a significant difference (P ¼ .06) between the treated and control tendons, with the treated tendons having a CSA of 0.64 cm2 and the control lesions having a CSA of 0.80 cm2. However, when the lesion size was taken as a percent of tendon CSA, treated tendon lesions were found to be 31% of that of tendon CSA, whereas control tendon lesions were 37% of tendon CSA, thus the difference was no longer significant (P ¼ .69). On ultrasound it was found that lesion-CSA measurement increased during the initial 42 days of the study (28 days into treatment period), and then stabilized (Fig. 1). There were no other significant differences in the ultrasound variables for treated and control tendons (Table 1).

DISCUSSION SDF tendonitis often requires a longer period of convalescence and has a higher rate of recurrence because as compared with the normal tendon the healed tendon has lesser tensile strength and elasticity.2,3 Many different therapies have been or are currently being investigated to treat this condition and to hopefully improve the quality of the repaired tissue.2,3,7,8,10,12,13 UBM, if effective, would be advantageous because it is a commercially available product from the bladder basement membrane that requires no special preparation or ex vivo expansion as it is acellular in structure. UBM has a proposed mechanism of action that involves the collagen, proteoglycans, glycoproteins, and growth factors of the extra-cellular matrix carrying out the following functions: to act as a bioscaffold in the repair of tissue ingrowth, to produce a local inflammatorylike response, and to recruit bone marrow-derived cells into the treated tissue.12,14,18

TW Wallis et al  Vol 30, No 7 (2010)

368

Table 1. Outcome parameters for SDF tendonitis in treated (UBM) and control (saline) groupsa Response Variable Lameness grade (AAEP Scale) Clinical parameters Palpation score Limb Circumference (cm) Tendon width (mm) Ultrasound parameters Lesion CSA (cm2) Lesion CSA as percentage of tendon CSA (%) Fiber score (0–4)b Echogenicity score (0–4)b Lesion length (cm) Gross pathology Gross length (mm) Gross width (mm) Gross depth (mm) Hemorrhage length (mm) Histology Tendon cell shape (8–32)b Tendon cell density (8–32)b Free hemorrhage (8–32)b Neovascularity (8–32)b Inflammatory cell infiltrate (8–32)b Collagen fiber linearity (4–16)b Collagen fiber uniformity (4–16)b Polarized collagen crimping (4–16)b Epitenon thickening (4–16)b Proximal sample total (6.2–24.9)b Proximal-middle sample total (6.2–24.9)b Middle-distal sample total (6.2–24.9)b Distal sample total (6.2–24.9)b Middle samples total (6.2–24.9)b Transition zone samples total (6.2–24.9)b Average total histology score (6.2–24.9)b

Treated

Control

P-value

1.21  0.11

1.44  0.11

.1

2.02  0.14 20.28  0.15 31.87  0.76

1.98  0.13 20.15  0.13 31.82  0.63

.97 .97 .12

0.64  0.06 31.58  2.97 2.55  0.18 2.38  0.18 8.05  0.30

0.80  0.07 36.93  3.26 2.58  0.18 2.48  0.18 8.13  0.31

.06 .69 .43 .83 .79

76.30  5.06 24.24  1.59 14.45  0.58 87.21  5.47

79.20  5.42 25.33  0.71 15.62  0.57 87.60  7.41

.7 .54 .17 .97

16.63  0.80 16.63  0.68 12.50  1.40 18.88  0.77 17.88  1.22 7.25  0.59 8.50  0.46 7.13  0.35 6.87  0.48 11.68  0.39 14.38  0.54 15.38  0.78 14.69  0.96 14.88  0.46 13.19  0.48 14.03  0.38

15.25  0.62 16.00  0.93 11.13  1.32 18.75  1.11 16.13  0.81 6.13  0.52 7.50  0.53 6.00  0.46 7.12  0.72 11.50  0.46 13.19  0.66 14.25  0.81 13.06  1.06 12.28  0.57 13.72  0.67 13.00  0.56

.2 .59 .49 .93 .25 .17 .18 .07 .78 .76 .19 .33 .28 .18 .23 .15

SDF, superficial digital flexor; UBM, urinary bladder matrix; AAEP, American Association of Equine Practitioners; CSA, cross-sectional area. a All values are reported as mean  standard error. Values for all clinical and ultrasound parameters represent mean values over the course of the study period, with measurements being taken on days 0 (baseline), 7, 14, 21, 28, 42, 56, 70, and 84. b Ranges given represent lowest and highest total possible score for the category. Lower numbers are closer to normal.

Although one clinical study suggested that UBM is an effective treatment for tendonitis in horses,12 no controlled clinical trial with histologic evaluation has been performed until now. On the basis of the results of the present study, UBM does not appear to be an effective treatment for collagenase-induced SDF tendonitis. No significant advantage or detriment was seen in the tendons treated with the product. However, the CSA of the tendon lesion as seen on ultrasound appeared to be smaller in the treated tendons as compared with the control tendons, and the control limbs had higher lameness scores as compared with the treated limbs. These differences were

however not statistically significant. In addition, no adverse reactions were seen in the treated limbs because of the use of UBM in treatment. There were some limitations in this study which should be considered. In the pilot studies, the lesions appeared to stabilize ultrasonographically after 2 weeks. However, in this study, the lesions continued to enlarge in both groups for 42 days, including the initial 28 days of the treatment period. This is an unexpected finding that has not been reported in previous studies, which might suggest that in our model the lesions failed to stabilize because of certain unknown reasons that extended the action of

TW Wallis et al  Vol 30, No 7 (2010)

369

0.8

promoted a quicker return to normalcy as compared with results from other clinical cases, thus suggesting substantial healing ultrasonographically at 60 days.12,14 However, on the basis of the results of this study, UBM did not significantly enhance healing of SDF tendonitis in this experimental collagenase model.

0.6

ACKNOWLEDGMENTS

0.4

This project was supported by funds obtained from ACell Europe, Alm Brand Insurance Company (Copenhagen, Denmark), and the Gail Holmes Equine Orthopaedic Research Center at Colorado State University.

1.4

Lesion CSA (cm^2)

1.2

Control Treated

1

0.2 0 0

14

28 Day

56

84

Figure 1. Ultrasonographic lesion sizes of SDF tendonitis in control and treated horses over the course of the treatment period. On ultrasound, it was found that the lesions that were created by injecting collagenase into the tendons continued to enlarge in both groups for 42 days, including the initial 28 days of the treatment period.

REFERENCES 1. Dyson SJ. Superficial flexor tendonitis in event horses, show jumpers, and dressage horses. In: Ross MW, Dyson SJ, eds. Diagnosis and management of lameness in the horse. Philadelphia, PA: WB Saunders; 2003:639–643. 2. Ross MW. Management of superficial digital flexor tendonitis. In: Ross MW, Dyson SJ, eds. Diagnosis and management of lameness in the horse. Philadelphia, PA: WB Saunders; 2003:625–639. 3. Davis CS, Smith RK. Diagnosis and management of tendon and

collagenase, other endogenous collagenases, or metalloproteinases of the matrix. Because in both groups the lesions enlarged for the same period, despite the fact that the control tendons only received a single intralesional saline injection, we suspected continued collagenase activity into the treatment period. However, other unidentified factors may have contributed to the lesion enlargement. Although the manufacturer recommends injecting 6 mL of the UBM intralesionally, a volume of only 4 mL was chosen for this study. This low volume was chosen on the basis of the projected lesion size from the pilot study for the following reasons: for maintenance of uniformity, and to avoid any possible reaction to the UBM that could occur as a result of overfilling of the lesion and subsequent subcutaneous leakage. However, in retrospect, the lesions were large enough for 6 mL treatments and possibly more. The manufacturer recommendations also suggested starting hand-walking within 24 hours of treatment. This was not followed in our model because of the bilateral and relatively severe nature of the lesions. However, similar to other types of intralesional treatments, exercise was begun at 14 days post treatment. Additionally, this collagenase model may not mimic clinical disease, as most of the tendon lesions created in this study appeared to be larger and more severe as compared with the lesions seen clinically. Furthermore, although any treatment for tendonitis is aimed at reducing the healing time, the 84-day study period may have been too short to have detected a significant difference between the treatments. This period was chosen to determine whether UBM

ligament disorders. In: Auer JA, Stick JA, eds. Equine surgery, 2nd ed. Philadelphia, PA: Saunders; 2006:1086–1110. 4. Smith RK, Webbon PM. The physiology of normal tendon and ligament. In: Proceedings of the Dubai International Equine Symposium. Dubai, UAE: Neyenesch Printers; 1996:55–81. 5. Smith RK. Pathophysiology of tendon injury. In: Ross MW, Dyson SJ, eds. Diagnosis and management of lameness in the horse. Philadelphia, PA: Saunders; 2003:616–628. 6. McIlwraith CW. Diseases and problems of tendons, ligaments and tendon sheaths. In: Stashak TS, ed. Adams’ lameness in horses. Philadelphia, PA: Lippincott Williams & Wilkins; 2002:459–644. 7. Dahlgren LA. Review of treatment options for equine tendon and ligament injuries: what’s new and how do they work. Proc Am Assoc Equine Pract 2005;51:376–382. 8. Dahlgren LA, Nixon AJ, Brower-Toland BD. Effects of betaaminopropionitrile on equine tendon metabolism in vitro and on effects of insulin-like growth factor-I on matrix production by equine tenocytes. Am J Vet Res 2001;62:1557–1562. 9. Foland JW, Trotter GW, Powers BE, Wrigley RH, Smith FW. Effect of sodium hyaluronate in collagenase-induced superficial digital flexor tendinitis in horses. Am J Vet Res 1992;53:2371–2376. 10. Henninger R. Treatment of superficial digital flexor tendinitis. Vet Clin North Am Equine Pract 1994;10:409–424. 11. Reef VB, Genovese RL, Davis WM. Initial long-term results of horses with superficial digital flexor tendinitis treated with intralesional baminoproprionitrile fumarate. Proc Am Assoc Equine Pract 1997; 43:301–305. 12. Mitchell RD. Treatment of tendon and ligament injuries with UBM powder (ACell-Vet). In: Proceedings of the 14th Annual American College of Veterinary Surgeons Symposium. Chicago, IL; 2004: 190–193.

370

13. Dahlgren LA, van der Meulen MC, Bertram JE, Starrak GS, Nixon AJ. Insulin-like growth factor-I improves cellular and molecular aspects of healing in a collagenase-induced model of flexor tendinitis. J Orthop Res 2002;20:910–919.

TW Wallis et al  Vol 30, No 7 (2010)

17. Badylak SF. The extracellular matrix as a scaffold for tissue reconstruction. Semin Cell Dev Biol 2002;13:377–383. 18. Badylak SF. Xenogeneic extracellular matrix as a scaffold for tissue reconstruction. Transplant Immunol 2004;12:367–377.

14. Badylak SF. Extracellular matrix as a scaffold for tissue engineering in

19. Dowling BA, Dart AJ, Hodgson DR, Rose RJ, Walsh WR. The effect

veterinary medicine: applications to soft tissue healing. Clin Tech

of recombinant equine growth hormone on the biomechanical prop-

Equine Pract 2004;3:173–181. 15. Badylak SF, Park K, Peppas N, McCabe G, Yoder M. Marrow-derived

erties of healing superficial digital flexor tendons in horses. Vet Surg 2002;31:320–324.

cells populate scaffolds composed of xenogeneic extracellular matrix.

20. Ross MW. Movement. In: Ross MW, Dyson SJ, eds. Diagnosis and

Exp Hematol 2001;29:1310–1318. 16. Badylak SF, Record R, Lindberg K, Hodde J, Park K. Small intestinal submucosa: a substrate for in vitro cell growth. J Biomater Sci Polym Ed 1998;9:863–878.

management of lameness in the horse. Philadelphia, PA: WB Saunders; 2003:60–73. 21. Reef VB, ed. Musculoskeletal ultrasonography. In: Equine diagnostic ultrasound. Philadelphia, PA: WB Saunders; 1998:39–186.