Introduction
Inflammation of the proximal aspect of the interosseus medius muscle, or suspensory ligament (SL), (ie, proximal suspensory desmitis) of one or both pelvic or thoracic limbs is a common cause of lameness of horses.1–4 Most horses lame because of proximal suspensory desmitis of one or both thoracic limbs become sound after a period of rest followed by controlled exercise,5 whereas most horses lame because of proximal suspensory desmitis of one or both pelvic limbs remain lame after the same treatment.1,6
Damage to the deep branch of the lateral plantar nerve (DBLPN) caused by compression has been theorized by some investigators to be responsible, at least in part, for the poor response of horses with proximal suspensory desmitis of a pelvic limb to rest.2,7,8 The DBLPN and its branches, the medial and lateral plantar metatarsal nerves, innervate the proximal aspect of the SL of the pelvic limb and lie between the SL and dense transverse fascia. The SL and the DBLPN and its branches are confined within a canal bounded by the transverse fascia, the plantar aspect of the third metatarsal bone, and the axial borders of the second and fourth metatarsal bones.2 The DBLPN and its branches may become compressed within this canal when the SL enlarges in response to inflammation, and compression of these nerves within this rigid enclosure may result in neuropathic pain, causing the horse to remain lame long after desmitis has resolved.2,7,8
Neurectomy of the DBLPN has been effective for many horses in eliminating chronic lameness caused by proximal suspensory desmitis of a pelvic limb, either by relieving pain in the SL or by alleviating neuropathic pain.2,3,7,8 Pauwels et al9, however, observed that the muscle fibers in the proximal aspect of a SL harvested from a pelvic limb of a horse 6 weeks after the ligament was denervated had histologic signs of neurogenic atrophy. They cautioned that reduced strength and elasticity of a denervated SL, caused by atrophy of the muscle fibers, coupled with loss of sensation in the ligament, may induce more ligamentous damage and increase the risk of catastrophic failure of the ligament.
A study10 by Lopez-Navarro et al showed that denervating the proximal aspect of the SL of both thoracic limbs of eight horses resulted in atrophy of muscle fibers and infiltration of the ligaments with connective and adipose tissue. One of the denervated ligaments of each horse in that study had been previously injected with collagenase to induce desmitis, and changes typical of neurogenic atrophy were more pronounced in the SL in which inflammation had been induced than in the non-injected, denervated SL of the contralateral thoracic limb. A study11 by Dyson et al (2016) found that horses with naturally occurring desmitis of a SL of a pelvic limb had histological evidence of neurogenic atrophy. Based on the study10 by Lopez-Navarro et al, the study9 by Dyson et al and the study9 by Pauwels et al, we hypothesized that denervating the SL of the pelvic limb would cause the muscle fibers of ligament to develop morphological and histological signs of atrophy and that atrophy would be more pronounced in denervated ligaments in which inflammation had been induced with collagenase than in ligaments that were only denervated.
To test our hypotheses, we compared morphological and histological characteristics of the proximal aspect of equine SLs that had been injected with collagenase to induce desmitis and later denervated with those of the proximal aspect of the contralateral ligament that had undergone denervation only. We also compared morphological and histological characteristics of non-injected, denervated SLs with those of non-denervated SLs harvested from the pelvic limbs of horses similar in size and age to those of horses from which the denervated ligaments were harvested.
Materials and Methods
Animals
Five sound, healthy adult horses (1 Thoroughbred and 4 mixed-breed horses), 5 to 12 years old (mean ± SD, 8 ± 2.5 years), weighing 317 to 435 kg (mean, 370.4 ± 44.3 kg), and having no evidence of proximal suspensory desmitis of the pelvic limbs, as determined by physical and ultrasonographic examinations, supplied the 10 treatment ligaments (ie, denervated SLs [treatment groups A and B]) for this study. Five sound, healthy adult mixed-breed horses, 7 to 10 years old (mean, 8 ± 1.1 years), weighing 292 to 415 kg (mean, 341.8 ± 52.4 kg), and having no physical or ultrasonographic evidence of proximal suspensory desmitis of the pelvic limbs supplied 10 control (ie, nondenervated) SLs. Horses received hay and water ad libitum and were housed individually in 4-m2 stalls.
The study was approved and supervised by the Institutional Committee for the Care and Use of Experimental Animals of the Faculty of Veterinary Medicine of the National Autonomous University of Mexico (approval No. MC-2014-18). A timeline of the project is provided (Supplementary Figure S1).
Induction of desmitis
Desmitis was induced in the proximal aspect of the SL of one randomly selected pelvic limb of each of the 5 horses supplying the treatment ligaments. To induce desmitis, the proximoplantar aspect of the randomly selected pelvic limb was prepared for sterile injection, the horse was sedated with xylazine HCl (0.5 mg/kg, IV), and skin at the proposed site of injection was desensitized by administering local anesthetic solution subcutaneously. A solution (0.3 mL) containing 1,500 U of filter-sterilized collagenase type 1 (Sigma-Aldrich Co) in sterile water was injected through a 22-gauge hypodermic needle into the SL, using ultrasound guidance, 4 cm distal to the head of the fourth metatarsal bone, and a sterile bandage was applied to the distal portion of the treated limb. The horses were administered phenylbutazone (2.2 mg/kg, IV) after injection and once daily for next 3 days. Bandages were changed once daily for 10 days, after which time the limbs were left unbandaged. The horses were confined to their stall, without exercise, for 2 weeks after injection, and after this time, they were walked in hand for 15 minutes 3 times daily for 4 weeks.
Neurectomy
The SLs of both pelvic limbs of all 5 horses supplying the treatment ligaments (ie, treatment groups A and B) were denervated 42 days after inducing desmitis by excising a segment of the DBLPN. To denervate the SLs, the horses were sedated with xylazine (1.1 mg/kg, IV), and anesthesia was induced with a combination of ketamine HCL (2.2 mg/kg, IV), diazepam (0.05 mg/kg, IV), and guaifenesin (50 mg/kg, IV). Anesthesia was maintained with a mixture of oxygen and isoflurane, administered through a circle system. Surgery was performed as previously described,7 and a sterile bandage was applied to each hock. Horses were administered gentamicin sulfate (6.6 mg/kg, IV) and phenylbutazone (2.2 mg/kg, IV) before surgery, and phenylbutazone, was administered twice daily, orally (2.2 mg/kg) for 3 days after surgery. The horses were walked twice daily for 10 minutes, and the pelvic limbs were rebandaged every 3 to 4 days, until skin sutures were removed at 15 days after surgery. Tissue excised was confirmed histologically to be nerve.
Morphologic and histological examinations of the SLs
The SLs from all 10 horses were harvested for morphological and histological examination after the horses were euthanized. Those horses supplying the treatment ligaments were euthanized at 120 days after being denervated. The 5 SLs that had been only denervated were assigned to treatment group A. The 5 contralateral SLs that had been injected with collagenase 162 days earlier (and denervated 42 days later) were assigned to treatment group B. The control ligaments consisted of 10 SLs harvested from 5 horses that had not undergone neurectomy of the DBLPN. Treatment ligaments (ie, treatment groups A and B) were compared morphologically and histologically with the non-denervated ligaments (control ligaments). For this comparison, each treatment ligament was compared with the mean values obtained from all the control ligaments from the pelvic limbs of the control horses. The noninjected, denervated ligaments (treatment group A) were compared morphologically and histologically with the contralateral, collagenase-injected, denervated ligaments (treatment group B).
The SLs of all horses were harvested immediately after the horses were euthanized by dissecting the ligament from its origin on the third metatarsal bone and transecting its branches distally at the attachment of the branches to the proximal sesamoid bones. Five 0.5-cm long slabs were sectioned at 1-cm intervals from the proximal 7.5 cm of each ligament, beginning 1 cm distal to the proximal aspect of the origin of the ligament. The mediolateral (width) and dorsoplantar (thickness) dimensions of each of these slabs were measured once with a stainless-steel Vernier caliper (Pretul-Truper; CDMX) immediately after the ligaments were harvested and divided into slabs, by one of the investigators (JAG), who was blinded to which of the ligaments had been injected with collagenase 162 days earlier.
The slabs were placed into tissue cassettes, immersed in neutral buffered 10% formalin for 24 hours, rinsed with tap water for two minutes, and placed in PBS solution for another 24 hours. The slabs were then dehydrated by placing them in increasingly concentrated baths of isopropyl alcohol (70%, 80%, 96%, and 100%), for 2 hours at each concentration, cleared in benzene for 1 hour, by using a TP 1020 Histokinette tissue processor (Leica. Glattbrugg) and embedded in paraffin wax.
A 4-μm-thick section was cut transversely from each of the 5 slabs harvested from each ligament by using a low-profile, stainless-steel blade (Accu-Edge) in a RM2125 RT microtome (LEICA Microsystems), mounted on a slide, and stained with H&E. The 2 slabs with the least artifacts were identified from these slides, and a 4-μm-thick, transverse section was cut from each of these 2 slabs, mounted on a slide, and stained with Masson trichrome stain. Three areas containing a high concentration of muscle fibers of one randomly chosen area were photographed with a DP70 digital camera (Olympus) mounted on a Bx41 microscope (Olympus). Each of the 3 photographs was examined at 40x magnification, and the mean cross-sectional area (CSA) of muscle fibers was calculated for each ligament by using ImageJ 1x.-Software (National Institute of Health). The mean cross-sectional area of muscle fibers contained within each of the 3 photomicrographs was calculated for each of the 10 control ligaments (30 images) and for each of the 10 denervated ligaments (30 images). Two pathologists (LR and RD), blinded to the identity of each ligament, evaluated the histological appearance of the sections in the photomicrographs.
Data analyses
The data were analyzed using SAS procedure PROC GLM in SAS/STAT (version 9.1.3). The dimensional measurements (ie, width, thickness, and CSA) were analyzed using the Shapiro Wilk test to determine normality distribution of the variables for the groups (ie, control ligaments and treatment ligaments [groups A and B]). The Levene test for variance ratio was performed to determine if the variances of the outcome variable were equal between groups, and a 2-sample t test for unequal variances was used to compare each dimensional measurement between groups. A paired t test was performed to compare each dimensional measurement between the 2 treatment groups (group A, non-injected, denervated ligaments; group B, collagenase-injected, denervated ligaments). A P value < 0.05 was selected as the threshold for significance for all statistical comparisons. All summary statistics in tables were reported as median and range due to the small sample size (n = 5) per group. Thickness and width were each reported as the mean of 5 measurements, and the CSA of muscle fibers contained within the 3 photomicrographs was reported as an average of three measurements.
Results
The mean age [8 years (range, 5 to 12 years)] and mean weight [370 kg (range 317 to 435 kg)] of the 5 horses from which the non-denervated SLs (ie, control ligaments) were harvested did not differ significantly from the mean age [ 8 years (range 7 to 10 years)] and weight [342 kg (range 292 to 415 kg)] of the 5 horses from which the denervated ligaments (ie, treatment ligaments) were harvested. All sex categories were represented in each of the two groups of horses supplying the ligaments. The thickness and width of the nondenervated SLs appeared grossly to be greater than that of the denervated ligaments, and the nondenervated ligaments were subjectively easier to cut than were the denervated ligaments during harvesting and sectioning. Neither the differences in measurements made with the caliper (ie, thickness and width) nor the differences in mean CSA of the muscle fibers, calculated from photomicrographs using imaging software between the right and left SLs among the 5 horses from which the non-denervated (control) ligaments were harvested, were statistically significant (Table 1).
Mean (range) values for paired comparisons of the suspensory ligaments (SLs) of the right and left pelvic limbs of the nondenervated (control) ligaments (n = 5) of healthy horses.
Measurement | Left SL | Right SL | P value |
---|---|---|---|
Width of body (mm) | 16.40 (15.86–17.38) | 16.46 (15.92–17.40) | 0.54 |
Thickness of body (mm) | 11.78 (11.02–11.85) | 11.34 (10.99–11.74) | 0.69 |
CSA of muscle fibers (mm2) | 3.012 (2.417–3.610) | 3.019 (2.489–3.617) | 0.64 |
Collagenase injected into the proximal aspect of the SLs caused the injected ligaments to enlarge and the horses to become chronically lame on that limb, but all horses became sound after neurectomy of the DBLPN of both pelvic limbs. At day 120 days after neurectomy (ie, the day the horses supplying the treatment ligaments were euthanized), the mean thickness, determined using a caliper of the collagenase-injected, denervated ligaments (treatment group B) was significantly (P = 0.03) less than that of the noninjected, denervated ligaments (treatment group A; Table 2). Similarly, the mean CSA of muscle fibers of the collagenase-injected ligaments (treatment group B), calculated from photomicrographs using imaging software, was significantly (P = 0.003) than the mean CSA of muscle fibers of the non-injected, denervated ligaments (treatment group A). The mean width of the collagenase-injected, denervated ligaments (treatment group B), as determined with a caliper, did not differ significantly (P = 0.10) from that of the noninjected, denervated ligaments (treatment group A).
Mean (range) values for comparisons between noninjected, denervated equine SLs (ie, treatment group A ligaments; n = 5) and collagenase-injected, denervated equine SLs (ie, treatment group B ligaments; 5) 120 days after ligaments in each group were denervated.
Measurement | Noninjected | Collagenase-injected | P value |
---|---|---|---|
Width of body (mm) | 14.84 (12.97–16.71) | 13.59 (11.83–15.35) | 0.10 |
Thickness of body (mm) | 10.73 (9.91–11.54) | 9.68 (8.27–11.10) | 0.03 |
CSA of muscle fibers (mm2) | 0.233 (0.161–0.305) | 0.215 (0.150–0.280) | 0.002 |
The mean width and thickness of the collagenase-injected, denervated ligaments (treatment group B) as determined with a caliper, were significantly (P = 0.002 and 0.024, respectively) than that of the nondenervated ligaments (control group; Table 3). The mean CSA of muscle fibers of collagenase-injected, denervated ligaments (treatment group B), determined from the photomicrographs, was significantly (P < 0.001) less than that of the non-denervated ligaments (control group).
Mean (range) values for comparisons between collagenase-injected, denervated equine SLs (ie, treatment group B ligaments; n = 5) and nondenervated equine SLs (ie, control group ligaments; 5).
Measurement | Collagenase-injected, denervated | Nondenervated | P value |
---|---|---|---|
Width of body (mm) | 13.59 (11.83–15.35) | 16.43 (15.86–17.40) | 0.002 |
Thickness of body (mm) | 9.68 (8.27–11.10) | 11.42 (10.99–11.85) | 0.024 |
CSA of muscle fibers (mm2) | 0.215 (0.150–0.280) | 3.017 (2.417–3.617) | < 0.001 |
The mean width and CSA of muscle fibers of the noninjected, denervated ligaments (treatment group A) were also significantly (P = 0.04 and < 0.001, respectively) less than that of non-denervated ligaments (control ligaments; Table 4). The mean thickness of the noninjected, denervated ligaments (treatment group A) did not differ significantly (P = 0.07) from that of nondenervated ligaments (control ligaments). The mean CSA of muscle fibers of all denervated ligaments (treatments A [0.233 mm2] and B [0.215 mm2]) was significantly (P = 0.002 for both comparisons) smaller than that of all the nondenervated ligaments (control ligaments [3.017 mm2).
Mean (range) values for comparisons between noninjected, denervated equine SLs (treatment group A ligaments; n = 5) and nondenervated equine SLs (control ligaments).
Measurement | Noninjected, denervated | Nondenervated | P value |
---|---|---|---|
Width of body (mm) | 4.84 (12.97–16.71) | 16.63 (15.86–17.40) | 0.039 |
Thickness of body (mm) | 10.73 (9.91–11.54) | 11.42 (10.99–11.85) | 0.70 |
CSA of muscle fibers (mm2) | 0.233 (0.161–0.305) | 3.017 (2.417–3.617) | < 0.001 |
Muscle fibers in segments of the SLs of horses that had undergone neurectomy of the DBLPN were atrophied and varied greatly in diameter (Figures 1 and 2). Changes typical of muscular atrophy observed microscopically included necrotic and ringed and whorled muscle fibers; muscle fibers with nuclei located centrally, rather than peripherally, as in normal muscle fibers; nuclear rowing; clusters of nuclei; infiltration of the ligament with fat and connective tissue; and fibrosis of the endomysium and perimysium. Two pathologists (LR and RD), blinded to the identity of the ligaments from which the photomicrographs were obtained, were unable to discern which of the photomicrographs were those of treatment group A ligaments and which were those of treatment group B ligaments.
Discussion
Our study showed unequivocally that the proximal aspect of the SL of the pelvic limb atrophies when the DBLPN is excised. The results of our study were similar to those of Lopez-Navarro et al,10 who found that neurectomy of the deep branch of the lateral palmar nerve resulted atrophy of the proximal aspect of the SL of the thoracic limb.
The SL of the pelvic limbs consists of dense collagenous tissue, small patches of fat, water, and varying amounts of muscle fibers.9,12–14 Wilson et al found that muscle accounts for about 12% (11.8 ± 4.2) of the CSA of the entire SL, including the branches.14 These muscle fibers, which are arranged in 2 longitudinal, parallel bundles, are concentrated predominately in the proximal third of the ligament and can account for up to 41% of the CSA of this portion of the ligament.13 The muscle fibers of the proximal aspect of the SL are 95% slow-twitch (ie, type I) and, therefore, have a relatively high resistance to fatigue.9,13 The short length and arrangement of these muscle fibers create a high potential for producing force, indicating that the SL of the horse is highly active.
Horses are commonly treated for lameness caused by proximal suspensory desmitis of the pelvic limb by neurectomy of the DBLPN of the affected limb or limbs,3,8 but the effects of denervation on the muscle fibers of the SL have not been well investigated. The innervation to the proximal aspect of the pelvic SL, provided by the DBLPN and its branches, provides the trophic factors necessary maintain the normality of the muscle fibers.9,15 The degenerative morphological changes observed in the SL after neurectomy of the DBLPN likely decreases the ligament’s load-bearing capacity by decreasing the ligament’s strength, elasticity, and ability to contract.9 Neurectomy of the DBLPN may increase the risk of catastrophic failure of the SL, and studies are needed to determine the clinical consequences of the degenerative changes produced by denervating this ligament.
Muscular atrophy was greater in the denervated ligaments injected with collagenase than in those denervated ligaments not injected with collagenase, as determined by reduced thickness of the SL and by reduced CSA of the muscle fibers. In a similar study, Lopez-Navarro et al10 found, during examination of the proximal aspect of SLs of the thoracic limbs, of 8 horses, harvested 8 weeks after neurectomy and 16 weeks after one of the thoracic SLs of each horse was injected with collagenase, that the quantity of muscle in the SLs injected with collagenase was significantly less than that of denervated ligaments that had not been injected with collagenase.10 These investigators theorized, that swelling of the proximal aspect of the SL, caused by inflammation, results in compression neuropathy, which in turn, may result in some muscular atrophy.
We examined only the proximal aspect of the pelvic SL for morphological and histological changes after neurectomy because this is the area that contains the most musculature and is the area most commonly damaged when the pelvic SL is injured.1,3,7,16 An investigation exploring the fate of the small quantity of muscle found within the distal portion of the body and the branches of the pelvic SL after neurectomy of the DBLPN may help determine the extent to which the DBLPN and its branches innervate the distal portion of the body and the branches of the SL of the pelvic limb.
Based on our findings and those of Pauwels et al and Lopez-Navarro et al, we conclude that neurectomy of the DBLPN, as a treatment for horses with proximal suspensory desmitis of one or both pelvic limbs, may predispose the horse to re-injury of the SL. The clinical significance of denervating the proximal aspect of the SLs of the pelvic limbs should be investigated.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org
Acknowledgments
The authors declare that there were no conflicts of interest and that no third-party funding or support was received in connection with this study (or case) or the writing or publication of the manuscript.
References
- 1. ↑
Dyson S. Proximal suspensory desmitis in the hindlimb: 42 cases. Br Vet J. 1994;150(3):279–291.doi: 10.1016/S0007-1935(05)80008-9
- 2. ↑
Dyson S. Proximal suspensory desmitis in the forelimb and the hindlimb. Proceedings of the 46th Annual Meeting of the American Association of Equine Practitioners. American Association of Equine Practitioners; 2002:137–142. doi:10.3920/978-90-8686-577-2
- 3. ↑
Dyson SJ, Murray R. Management of hindlimb proximal suspensory desmopathy by neurectomy of the deep branch of the lateral plantar nerve and plantar fasciotomy: 155 horses (2003–2008). Equine Vet J. 2012;44(3):361–367.doi: 10.1111/j.2042-3306.2011.00445.x
- 4. ↑
Singer ER, Barnes J, Saxby F, Murray JK. Injuries in the event horse: training versus competition. Vet J. 2008;175(1):76–81. doi:10.1016/j.tvjl.2006.11.009
- 5. ↑
Dyson S. The suspensory apparatus, In: Rantanen N, McKinnon A, eds. Equine Diagnostic Ultrasonography. Williams & Wilkins; 1998:447–473.
- 6. ↑
Hewes CA, White NA. Outcome of desmoplasty and fasciotomy for desmitis involving the origin of the suspensory ligament in horses: 27 cases (1995–2004). J Am Vet Med Assoc. 2006;229(3):407–412. doi:10.2460/javma.229.3.407
- 7. ↑
Tóth F, Schumacher J, Schramme M, Holder T, Adair HS, Donnell RL. Compressive damage to the deep branch of the lateral plantar nerve associated with lameness caused by proximal suspensory desmitis. Vet Surg. 2008;37(4):328–335. doi:10.1111/j.1532-950X.2008.00385.x
- 8. ↑
Bathe A. Neurectomy and fasciotomy for the surgical treatment of hindlimb proximal suspensory desmitis. Proceedings of the 40th British Equine Veterinary Association Congress. British Equine Veterinary Association; 2001;118.
- 9. ↑
Pauwels FE, Schumacher J, Mayhew G, Van Sickle DC. Neurectomy of the deep branch of the lateral plantar nerve can cause neurogenic atrophy of the muscle fibres in the proximal part of the suspensory ligament (M. interosseous III). Equine Vet J. 2009;41(5):508–510. doi:10.2746/042516409x435629
- 10. ↑
Dyson S, Murray R, Pinilla MJ. Proximal suspensory desmopathy in hindlimbs: a correlative clinical, ultrasonographic, gross postmortem and histological study. Equine Vet J. 2017;49(1):65–72. doi:10.1111/evj.12563
- 11. ↑
Lopez-Navarro G, Trigo TF, Rodriguez-Monterde A, et al. Histological changes in the proximal aspect of the suspensory ligament after neurectomy of the deep branch of the lateral palmar nerve of horses with induced proximal suspensory desmitis. Vet J. 2017;227:46–48.
- 12. ↑
Schramme M, Josson A, Linder K. Characterization of the origin and body of the normal equine rear suspensory ligament using ultrasonography, magnetic resonance imaging, and histology. Vet Radiol Ultrasound. 2012;53(3):318–328. doi:10.1111/j.1740-8261.2011.01922.x
- 13. ↑
Soffler C, Hermanson JW. Muscular design in the equine interosseus muscle. J Morphol. 2006;267(6):696–704. doi:10.1002/jmor.10433
- 14. ↑
Wilson DA, Baker GJ, Pijanowski GJ, Boero MJ, Badertscher RR II. Composition and morphologic features of the interosseous muscle in Standardbreds and Thoroughbreds. Am J Vet Res. 1991;52(1):133–139.
- 15. ↑
Valentine BA, McGavin MD. Skeletal muscle. In: McGavin MD, Zachary FZ, eds. Pathologic Basis of Veterinary Disease. 4th ed. Mosby Elsevier; 2007:991–993.
- 16. ↑
Souza MV, van Weeren PM, van Schie HT, van de Lest CHA. Regional differences in biochemical, biomechanical and histomorphological characteristics of the equine suspensory ligament. Equine Vet J. 2010;42(7):611–620. doi:10.1111/j.2042-3306.2010.0089.x