In dogs, signs of pain and lameness are generally more noticeable in thoracic limbs than in pelvic limbs, possibly owing to the disproportionate distribution of weight bearing between the thoracic and pelvic limbs, which may accentuate clinical signs. Signs of pain are frequently localized to the shoulder region of dogs, but that pain may originate from various sources including primary bone lesions (eg, osteosarcoma of the proximal aspect of the humerus) and lesions of supporting soft tissues or be referred from elbow joint disease or nerve root signature from primary spinal cord lesions. Supraspinatus tendon injury as a potential source of shoulder joint pain has garnered considerable interest from the veterinary orthopedic and rehabilitation communities.1–11 Clinical signs and diagnostic findings for dogs with supraspinatus tendon injury have been largely extrapolated from human medicine despite the fact that the morphology and loading forces of the shoulder joint differ substantially between dogs and humans.
In humans, the glenoid labrum (rotator cuff) is comprised of the tendons of 4 muscles (supraspinatus, infraspinatus, subscapularis, and teres minor muscles; ie, cuff tendons), which form a cuff over the head of the humerus and help to lift and rotate the arm in addition to stabilizing the shoulder joint.12 In dogs, although the shoulder joint is comprised of tendons of the same 4 muscles, those tendons do not form a proper cuff as they do in humans.12 In humans, degeneration and tearing of the supraspinatus tendon leads to shoulder joint instability and can be caused by several mechanisms such as chronic microtrauma, chronic degeneration, chronic impingement, acute avulsion, dislocation, iatrogenic injury, and nerve damage.13–17 Similar mechanisms likely lead to supraspinatus tendon degeneration and tearing in dogs. In humans, degeneration of the rotator cuff is exacerbated by specific events or conditions such as smoking, type of occupation, or participation in overhead sports18,19; however, specific conditions that predispose dogs to injury of the supraspinatus tendon have yet to be identified. Weakening of cuff tendons, including the supraspinatus tendon, results in loss of normal shoulder joint function and stability and signs of pain. In humans, weakening of the cuff tendons is characterized by pain that is exacerbated by overhead activity and muscle weakness. In dogs, injury to the tendons of the shoulder joint is characterized by nonspecific signs of pain localized to the cranial aspect of the shoulder area during flexion and extension of the joint.3,9 Because dogs cannot perform provocative maneuvers reliably or be instructed to isolate a specific muscle for testing, veterinarians often rely on gross localization of signs of pain to a specific region followed by diagnostic imaging of that region to arrive at a diagnosis.
Ultrasonography and MRI are considered the most clinically useful diagnostic imaging modalities for diagnosis of supraspinatus tendinosis in dogs.1,4,10 On MRI sequences of a 4-year-old, 18-kg dog with concomitant supraspinatus tendinosis and biceps brachii tendon displacement, the supraspinatus tendon appeared as a large mass-like insertion approximately 5 to 7 mm thick and had an increased signal intensity relative to that for the biceps brachii tendon on T2-weighted and other fluid-sensitive pulse sequences, which was attributed to an increase in local water content.11 Review of MRI sequences for older (age, 7 to 11 years) large-breed (weight, 32 to 50 kg) dogs without supraspinatus tendinosis revealed that the supraspinatus tendon generally had a hypointense signal.11 In other studies,3,9 the supraspinatus tendon of dogs with supraspinatus tendinopathy, particularly those with noncalcifying tendinopathy, appeared enlarged and had an increased signal intensity on MRI fluid-sensitive pulse sequences. The dogs evaluated in those 2 studies3,9 varied in terms of both age and size and represented fairly small breeds such as terriers and spaniels. Diagnosis of supraspinatus tendinopathy by evaluation of MRI sequences can be challenging because the supraspinatus tendon of the clinically normal contralateral limb of affected dogs and those of nonlame dogs often have similar appearances.3,9 During evaluation of MRI sequences of fresh cadaveric canine limbs obtained to determine the optimal protocol for quantitative imaging of the shoulder joint, our research group likewise observed that the supraspinatus tendons of nonlame 2-year-old sexually intact male Beagles had imaging characteristics similar to those reported for diseased tendons. Specifically, the supraspinatus tendons of those nonlame dogs had a trilaminar appearance with a higher signal intensity at the center than at the outer margins. The purpose of the study reported here was to characterize the MRI and histologic features of the supraspinatus tendon in nonlame dogs. Our hypotheses were that the insertion (tendinous portion) of the supraspinatus muscle of nonlame dogs would appear as a thick trilaminar structure with a hyperintense signal intensity in the center relative to the signal intensity of the outer margins on MRI images, and that the variation in the MRI signal intensity would be associated with the tissue composition of the different portions of the tendon.
Materials and Methods
Animals
The study was approved by the Cornell University and Hospital for Special Surgery Institutional Animal Care and Use Committees under the auspices of a tissue-sharing protocol (ie, all dogs were part of an unrelated study approved by the institutional animal care and use committee). The cadavers of 7 (ie, 14 shoulder joints) 2-year-old sexually intact male Beagles were obtained immediately after euthanasia. Review of in-house veterinary records indicated that all dogs had a normal gait and no history of lameness prior to euthanasia. The cadavers were stored at 4°C and underwent MRI within 72 hours after death.
MRI
Both shoulders of each cadaver were scanned with a 3-T scannera and 8-channel phased-array wrist coil.b Three-plane proton density–weighted FSE and single-plane short tau inversion recovery sequences were obtained for all 14 shoulders. Additional fluid-sensitive pulse (T2-weighted, T2-weighted spectral selective fat suppression, multiplanar gradient recalled acquisition in the steady state, and spoiled gradient recalled acquisition) sequences were obtained for 2 cadavers (ie, 4 shoulder joints) for comparison with results obtained by use of similar sequences in the veterinary literature. The parameters for all MRI sequences performed were summarized (Appendix).
All MRI images were reviewed by a board-certified veterinary radiologist (SLP) with 9 years of experience in performing and reviewing small animal MRI sequences. Images were viewed on standard viewing softwarec and evaluated for the presence of disrupted fibers or undermining of the insertion of the supraspinatus tendon, bone marrow edema of the greater tubercle of the humerus, and edema along the myotendinous junction, all of which are suggestive of supraspinatus tendon injury.20,21 For each shoulder joint, the superficial margin, central substance, and deep margin of the supraspinatus tendon were measured at its thickest portion on the sagittal midline image, and central substance-to-superficial margin and central substance-to-deep margin thickness ratios were calculated. Three-dimensional data were available for 11 of the 14 shoulders, and 3-D images were imported into an open-source segmentation software programd so that automatic contour segmentation of the supraspinatus tendon insertion could be performed and its signal intensity and volume measured. Stereolithography mesh files were generated and imported into a plotting software programe for presentation purposes. All 3-D analyses were performed by 1 investigator (SLP). Mean signal intensities were calculated for all voxels representing the supraspinatus tendon and for a comparable volume of the triceps brachii muscle, which were used to compute the supraspinatus tendon-to-triceps brachii muscle signal intensity ratio. Muscle quality of the supraspinatus muscle was assessed by means of a scoring system developed by Goutallier et al.22
Histologic examination
Following completion of MRI scanning, all shoulder joints were dissected from the cadavers and fixed in neutral-buffered 10% formalinf for a minimum of 2 weeks. Then, the specimens were further cut down to size with bone cutters and placed in an acid decalcification solution.g Decalcification of specimens was determined by chemical testing, which included 5 mL of decalcifying fluid combined with 1 mL 5% ammonium oxalate, and confirmed radiographically. Following decalcification, specimens were washed in water for 4 to 6 hours and then sectioned into 3-mm-thick slices in the sagittal or transverse plane. The specimens were processedg in a routine manner and placed in paraffin blocks. The blocked specimens were slicedh into 6-μm–thick sections. Each section was stained with H&E, Alcian blue (pH, 2.5), periodic acid–Schiff, a combination of Alcian blue (pH, 2.5) and periodic acid–Schiff (with and without hyaluronidase), safranin-O, Gomori trichrome, or picrosirius red stain.
All stained sections were examined by a board-certified veterinary pathologist (BGC) with 10 years of experience. Light microscopyi was used to evaluate sagittally sliced tendon specimens for collagen fiber density and orientation, overall thickness, relationship with surrounding structures, vascularity, matrix quality, and cellularity.
Results
MRI characteristics of the supraspinatus tendon
For all shoulder joints evaluated, there was no MRI evidence of joint pathology or edema at the supraspinatus musculotendinous junction, and the quality of the supraspinatus muscle was subjectively normal. The supraspinatus tendon had a trilaminar appearance on all MRI images regardless of the pulse sequence acquisition and parameter settings used (Figure 1). Specifically, on proton-density FSE images, the supraspinatus tendon was comprised of 3 definable layers (a thin hypointense band along the superficial margin, a thick hyperintense central substance, and a poorly demarcated deep margin that was hypointense relative to the central substance but not as hypointense as the superficial margin; Figure 2). On sagittal midline images of the supraspinatus tendon, the mean ± SD thickness was 0.7 ± 0.1 mm for the superficial margin, 5.5 ± 0.5 mm for the central substance, and 0.6 ± 0.1 mm for the deep margin. The mean ± SD central substance-to-superficial margin and central substance-to-deep margin thickness ratios were 8.4 ± 1.2 and 9.0 ± 0.9, respectively. The mean ± SD ratio for the mean signal intensity of the supraspinatus tendon relative to that for triceps brachii muscle was 1.3 ± 0.2 (range, 0.9 to 1.6). The mean ± SD volume of the supraspinatus tendon was 445 ± 20 mm3 (range, 422 to 480 mm3). For all shoulder joints evaluated, the insertion of the supraspinatus tendon was wide and crossed medially on the midline plane formed by the long axis of the scapula as previously described.23
Representative sagittal proton-density FSE (A), multiplanar gradient recalled acquisition (B), short tau inversion recovery (C), T2-weighted FSE and T2 with frequency-selective fat suppression (D), T2-weighted FSE with frequency-selective fat suppression (E), and spoiled gradient recalled echo (F) images of the shoulder joint of a nonlame 2-year-old sexually intact male Beagle cadaver. Notice that the supraspinatus tendon (arrows) has a trilaminar appearance and a hyperintense signal at its insertion on the greater tubercle of the humerus on all images.
Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.836
Representative transverse (A) and sagittal (B) proton-density FSE images of the supraspinatus tendon of a nonlame 2-year-old sexually intact male Beagle cadaver. The areas outlined in panels A and B have been magnified in panels C and D, respectively. Notice that the collagen-dense fibers of the superficial (arrowhead) and deep (white arrows) margins of the supraspinatus tendon have a hypointense signal, whereas the thick water-rich central substance (double-headed black arrow) has a hyperintense signal.
Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.836
Histologic features of the supraspinatus tendon
Histologic evaluation of the supraspinatus tendon revealed that it had 3 distinct layers or zones. The superficial zone varied in thickness from approximately 35 to 100 μm and consisted of loose fibrovascular connective tissue or a portion of an overlying muscle, which transitioned to a layer (approx 300 to 400 μm thick) of dense parallelly arranged tendon fascicles with little noncollagenous matrix. That zone transitioned to the largest layer of the tendon, which consisted of a 3- to 4-mm thick region of matrix-rich connective tissue that contained parallel or haphazardly arranged collagen bundles (diameter, 6 to 10 μm). The deepest portion of the tendon consisted of an approximately 300-μm–thick layer of dense collagenous fibers that transitioned to loose fibrovascular and adipose connective tissue at the shoulder joint capsule. Near its insertion on the greater tubercle of the humerus, the composition of the connective tissue matrix became more collagenous. The supraspinatus tendon measured approximately 5 mm at its widest point in the sagittal plane.
At the insertion of the supraspinatus tendon on the greater tubercle of the humerus, the central substance transitioned from an abundant mucinous and cartilaginous matrix to fascicular collagenous fibers that had a fairly linear arrangement (Figure 3). Those collagen fibers connected to the bone in an interwoven pattern of nonmineralized fibrocartilage that mineralized as it became associated with the bone. The proximal attachment of the tendon contained the most fibrocartilage, whereas the distal attachment was fibrous and eventually blended with the periosteum of the greater tubercle via small collagenous fibers.
Representative sagittal spoiled recalled gradient echo image of the insertion of the supraspinatus tendon on the greater tubercle of the humerus in a nonlame 2-year-old sexually intact male Beagle cadaver before (A) and after magnification (B) and photomicrographs of a sagittally sliced section of the tendon insertion at its most distal (C) and proximal (D) aspects that correspond to the areas highlighted in panel B. Notice that the distal aspect of the tendon insertion (bracket) is comprised primarily of collagenous fibers, whereas the proximal aspect of the tendon insertion (arrows) is comprised of a cartilaginous matrix. Panels C and D—H&E stain; bar = 400μm.
Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.836
Examination of tendon specimens that were prepared with various stains was useful for differentiating areas with high collagen and GAG contents (Figure 4). On H&E-stained specimens, the periphery of the tendon consisted of parallelly arranged eosinophilic fibrous fibers, whereas the central substance consisted of a basophilic matrix with hypocellular stroma that contained a few haphazard fibrocytic cells intermixed with large chondrocytic cells. For specimens stained with a combination of Alcian blue and periodic acid– Schiff stains, the matrix of the central substance had strong uptake of the Alcian blue stain, which was indicative of a high content of acid nonsulfated mucopolysaccharides (eg, hyaluronic acid). Subsequent treatment of those specimens with hyaluronidase resulted in a decrease in the stain intensity of the central substance, which confirmed the presence of hyaluronic acid and other GAGs. Specimens prepared with safranin-O stain revealed that the central substance also contained acid sulfated mucopolysaccharides, but the stain intensity suggested that the content of acid sulfated mucopolysaccharides was less than that of acid nonsulfated mucopolysaccharides. Specimens were prepared with picrosirius red stain to highlight the presence and orientation of fine and dense collagenous fibers within the tendon. Fiber orientation varied throughout the tendon, with the central substance containing interlacing bundles and haphazardly arranged fine collagenous fibers.
Representative photomicrographs of the matrix staining properties of the supraspinatus tendon of a nonlame 2-year-old sexually intact male Beagle cadaver. A—Photomicrograph of a sagittally sliced section of the distal aspect of the supraspinatus muscle and the insertion of its tendon on the articular surface of the greater tubercle of the humerus. Highlighted sections of the superficial margin and central substance of the tendon have been magnified (20×) above and below the panel, respectively. Notice that the periphery of the tendon is comprised of parallelly arranged eosinophilic fibrous tendon fibers, whereas the center of the tendon consists of a basophilic matrix characterized by hypocellular stroma with a few fibrocytic cells haphazardly intermixed with enlarged chondrocytes. H&E stain. B—Photomicrograph of a tissue specimen similar to that in panel A, which has been stained with a combination of Alcian blue (pH, 2.5) and periodic acid–Schiff stains to depict the presence of acid nonsulfated (simple) mucopolysaccharides. Notice that the acid nonsulfated mucopolysaccharide–rich central substance of the tendon is stained bright blue, whereas the dense collagenous fibers of the tendon periphery are not. C—Photomicrograph of a tissue specimen similar to that in panel A, which has been stained with Safranin-O stain to depict the presence of acid sulfated (complex) mucopolysaccharides. Similar to the acid nonsulfated mucopolysaccharides, most acid sulfated mucopolysaccharides are located in the central substance rather than the periphery of the tendon. D—Photomicrograph of a tissue specimen similar to that in panel A, which has been stained with picrosirius red stain to highlight the fine and dense collagenous fibers within the tendon. Notice that the orientation of collagenous fibers varies throughout the tendon, with the central substance containing interlacing bundles and haphazardly arranged fine collagenous fibers. Bar = 100 μm when present.
Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.836
Discussion
Results of the present study indicated that the supraspinatus tendon of nonlame dogs had a trilaminar appearance and variable signal intensity on MRI sequences, and such findings should not be interpreted as evidence of pathological lesions. Histologically, the staining qualities of the supraspinatus tendon of nonlame dogs was typical of those of other tendons at its periphery and enthesis. However, the central substance of the supraspinatus tendon appeared to have a fairly high fluid content with little dense collagen, which, when considered in combination with a strong Alcian blue (pH, 2.5) and mild-to-moderate safranin-O stain-positive matrix, suggested that it was primarily composed of connective tissue proteoglycans. That composition is distinct from that of articular cartilage, which has a mild Alcian blue and strong safranin-O stain-positive matrix. The fluid-rich central substance closely resembles the matrix of other tendons. In short, the supraspinatus tendon of dogs is essentially a fluid-rich matrix of GAGs sandwiched between 2 layers of dense collagen. The deep margin of the supraspinatus tendon has an indistinct transition to the shoulder joint capsule attachment. Differentiation of the precise GAG composition of the central substance matrix was not possible with the stains used for the processing of histologic specimens and was beyond the scope of this study.
The histologic results of the present study suggested that the fluid-rich central substance of the supraspinatus tendon was responsible for the high signal intensity observed on MRI fluid-sensitive pulse sequences. The MRI characteristics of the supraspinatus tendon were in contrast to those of tendons that have a more uniform composition. For example, the patellar tendon appears as a thick hypointense band of tissue on most MRI sequences and is comprised of tightly packed, parallelly arranged bundles of collagen fibers (Figure 5). The highly ordered collagen fibers cause a shortening of the T2 relaxation time and loss of signal on most standard MRI pulse sequences. The assumption that all tendons should be uniformly hypointense on MRI sequences would be true if all tendons had identical function and loading; however, the insertion of a tendon is influenced by regional mechanical forces, which dictate the organization and composition of the tendon matrix.24–26 On MRI images, the signal intensity of tendons is susceptible to the magic angle phenomenon, and that artifact was considered as a possible cause of the MRI characteristics for the canine supraspinatus tendons evaluated in the present study. However, the signal intensity of the haphazardly arranged collagen fibers of the central substance would not be affected by the magic angle phenomenon, and the highly ordered collagenous fibers of the superficial margin were not consistently positioned at 55° to the main magnetic field. Thus, we believe it unlikely that magic angle artifact interfered with interpretation of the MRI images of the supraspinatus tendon evaluated in this study. Positioning of the forelimb and shoulder joint might affect the presence or magnitude of magic angle artifact. On the basis of the results of this study, it appeared that the superficial margin of the canine supraspinatus tendon would be most susceptible to the magic angle phenomenon owing to the highly ordered nature of the collagen fibers in that region.27
Representative sagittal proton-density FSE image (A) and photomicrograph (B; H&E stain; bar = 5 mm) of the supraspinatus tendon (circle) of a nonlame 2-year-old sexually intact male Beagle cadaver. Notice that the tendon has a trilaminar appearance and the central substance has a hyperintense signal and more basophilic matrix than the adjacent superficial and deep margins. A representative sagittal proton-density FSE image (C) and photomicrograph (D; H&E stain; bar = 5 mm) of the patellar tendon (arrows) and caudal cruciate ligament (arrowhead) of a nonlame adult dog are provided for comparison with the supraspinatus tendon. Notice that both the patellar tendon and caudal cruciate ligament have a uniform low signal intensity on the MRI image and a histologic appearance characterized by eosinophilic layers of dense collagen fibers that is similar to that of the superficial and deep margins of the supraspinatus tendon.
Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.836
The MRI and histologic features of the canine supraspinatus tendons evaluated in the present study suggested that the central substance of the tendon had an abundant connective tissue matrix with high water and GAG contents, which likely reflects an adaptation to a wide range of motion that includes both compressive and shear forces. The collagen content of tendons is generally inversely proportional to the proteoglycan content, particularly in regions near attachments.25 In dogs, the type III collagen content at the insertion of the supraspinatus tendon is greater than that in other portions of the tendon.25 Type III collagen is believed to be more extensible than other types of collagen, and its abundance at the insertion of the supraspinatus tendon may be an evolutionary adaptation to the unique stresses that occur at the shoulder joints of dogs.25 Although type III collagen is disproportionately detected in damaged tenocytes relative to healthy tenocytes,27 it is important to understand the characteristics unique to the insertion of the canine supraspinatus tendon of dogs to prevent misdiagnosis of tendon trauma or degeneration. The central substance of the canine supraspinatus tendon is similar to, albeit distinct from, fibrocartilage or hyaline cartilage; it contains a matrix with a similar GAG composition but its staining intensity is paler, which is indicative of a lower GAG concentration. The lower GAG concentration of the central substance of the supraspinatus tendon relative to that of fibrocartilage or hyaline cartilage may allow for more movement and deformation. Cartilage is meant to resist compression, cartilaginous fibers are meant to resist tension, and fibrocartilage is meant to resist the combination of compression and tension (ie, shear forces). The central substance of the supraspinatus tendon is unique. Although it contains both fibrous and fibrocartilage tissue, especially at its insertion on the humerus, in dogs, the matrix of the remainder of the supraspinatus tendon has a fairly high fluid content, which is believed necessary to allow for a wide range motion.26
For a dog of 1 clinical report,11 a diagnosis of nonmineralizing supraspinatus tendinopathy was made on the basis of MRI findings and histologic results for focal biopsy specimens obtained from the suspected affected region. In that report,11 tissue specimens believed to have an abnormal MRI signal intensity were histologically characterized as having a pale basophilic matrix with loose interlacing narrow bundles of collagen fibers and a few chondrocytic cells. If those specimens were obtained from the central substance of the tendon and appropriate control specimens were unavailable for comparison purposes, those findings may have been incorrectly interpreted as evidence of mucinous or chondroid degeneration indicative of maladaptive or pathological lesions. Tendon damage or tearing should not be diagnosed in the absence of inflammation, neovascularization, fragmented fibers, hemorrhage, and fibrosis or scarring. Histologic evidence of mucinous or chondroid degeneration in focal biopsy specimens should be interpreted cautiously because the identification of well-organized structures during examination of the entire tendon might refute such a diagnosis. In the present study, the MRI and histologic characteristics observed for the supraspinatus tendons of 7 nonlame adult Beagles were similar to those described for dogs with shoulder joint abnormalities in other reports.4,10,11 For all dogs of the present study, both supraspinatus tendons transitioned from a densely arranged fibrous superficial margin to a loosely arranged mucocartilaginous central substance back to a densely arranged fibrous deep margin, with no evidence of pathological changes such as hemorrhage, torn fibers, fibrosis, or granulation tissue. Therefore, the MRI findings described for the supraspinatus tendons in this study should be considered normal. In humans, calcaneal tendons with histologic evidence of mucoid or chondroid degenerations also have deformation of the adjacent densely packed collagen bundles,28 a feature that was not observed in the canine supraspinatus tendons evaluated in this study. Additionally, hyaline cartilage was not present in the supraspinatus tendons evaluated in this study.
The supraspinatus tendons of nonlame dogs are similar to rotator cuff tendons of humans in that they have a central region rich in GAGs and an arrangement of collagen fibers that facilitates a wide range of movement while providing resistance to compressive forces.6 These features are unique to the supraspinatus tendon and, to our knowledge, had not been described for dogs prior to the present study. On the basis of the results of this study, we believe that a supraspinatus tendon with a widened insertion and central hyperintense signal on MRI fluid-sensitive pulse sequences should be considered normal in the absence of MRI evidence of other pathological lesions such as edema at the myotendinous junction, undersurface delamination of the tendon fibers, insertional edema of the greater tubercle of the humerus, and intraosseous ganglion cysts at the tendon footprint.
In dogs, unlike humans, mineralization of the supraspinatus tendon commonly develops and is often subclinical and bilateral. Dogs with clinical signs associated with supraspinatus tendon mineralization have been managed surgically with variable long-term success.29 Mineralization can develop at any location within the supraspinatus tendon and is best described as enthesitis when it develops at the insertion of the tendon on the greater tubercle of the humerus and enthesopathy when it spreads from the fibrocartilaginous portion of the tendon to the normal zone of calcification near its attachment to the bone and is eventually replaced with bone as a result of endochondral ossification.30 Distant from the enthesis, calcification of the supraspinatus tendon can develop subsequent to mineralization of degenerate collagen or fibrosis (degenerative calcification) or metaplasia of the tendon to fibrocartilage followed by mineralization (reactive calcification).30 Although similar, degenerative and reactive calcification processes are distinct from the enthesitis process, and the 3 processes should be viewed as separate conditions despite the fact that they have been compiled into the same category in multiple veterinary reports.29,31 Additionally, both degenerative calcification and reactive calcification have been described in conjunction with supraspinatus tendinopathy in dogs regardless of cause, pathogenesis, or pathological morphology.1,31
A limitation of the present study was the fact that both shoulder joints were evaluated for each of 7 canine cadavers with a uniform breed, age, and sex. It would have been ideal for the MRI scans to be performed on live dogs, but that was not feasible. The cadavers were stored in a chilled environment and scanned within 72 hours after euthanasia and did not undergo any freeze-thaw cycles. Autolysis of connective tissue is fairly slow in refrigerated beef during postmortem aging,32 and presuming canine tissue reacts similarly, significant proteoglycan and collagen loss within the supraspinatus tendon should have been minimal between euthanasia and MRI scanning for the cadavers of this study. Our research group has performed MRI on the shoulder joints on a small number of sexually intact male Beagles that were older than those evaluated in the present study. The MRI characteristics of the supraspinatus tendon for the older live dogs were similar to those for the 2-year-old canine cadavers of this study (Figure 6). Comparison of the MRI characteristics of supraspinatus tendons between old and young dogs is important because age-related changes occur in the tendons of horses.33 Additional in vivo and ex vivo MRI of the supraspinatus tendons and shoulder joints of dogs of various breeds and ages is necessary. Also, because tendon composition can be affected by hormones, the MRI characteristics of the supraspinatus tendon should be compared between male and female dogs that have and have not been neutered.
Representative sagittal proton-density FSE images of a shoulder joint of a nonlame 13-year-old sexually intact male Beagle obtained in vivo (A) and a nonlame 2-year-old sexually intact male Beagle cadaver (B). In both panels, notice that the supraspinatus tendon has a trilaminar appearance and that the superficial (arrow) and deep margins have a hypointense signal relative to the hyperintense signal of the central substance.
Citation: American Journal of Veterinary Research 79, 8; 10.2460/ajvr.79.8.836
In the present study, the MRI and histologic characteristics of the supraspinatus tendons of 7 nonlame 2-year-old sexually intact male Beagle cadavers were described. Many of those characteristics were similar to those described for dogs with confirmed or presumed shoulder joint pathology.4,11 Therefore, we concluded that, in dogs, a diagnosis of supraspinatus tendinosis should not be based solely on the presence of a hyperintense signal at the insertion of the tendon on the greater tubercle of the humerus on MRI fluid-sensitive pulse sequences. Additional MRI evaluation of the shoulder joints of sexually intact and neutered male and female dogs of various breeds and ages is necessary to confirm that histologically normal supraspinatus tendons have a trilaminar appearance with a thick central substance that is hyperintense on MRI images. Until those studies have been performed, the presence of other MRI criteria such as an edema pattern at the myotendinous junction or greater tubercle of the humerus, evidence of tendon avulsion at the greater tubercle of the humerus, and other evidence of joint degeneration can be used to increase the index of suspicion for supraspinatus tendinosis or tendinopathy.
Acknowledgments
The Hospital for Special Surgery MRI Laboratory receives institutional research support from GE Healthcare; however, GE Healthcare did not provide any financial support for this study.
ABREVIATIONS
| FSE | Fast spin echo |
| GAG | Glycosaminoglycan |
Footnotes
GE Healthcare, Waukesha, Wis.
Invivo, Gainesville, Fla.
Clear Canvas, Toronto, ON, Canada.
ITK-SNAP, version 3.6.0, Yushkevich P, Penn Image Computing and Science Laboratory, Department of Radiology, University of Pennsylvania, Philadelphia, Penn, and Gerig G, Scientific Computing and Imaging Institute, University of Utah, Salt Lake City.
Tecplot Focus 2013, Bellevue, Wash.
VWR, Radnor, Pa.
Triangle Biomedical Sciences, Durham, NC.
Shandon-Elliot, London, England.
Olympus America, Center Valley, Pa.
References
1. Canapp SO Jr, Canapp DA, Ibrahim V, et al. The use of adipose-derived progenitor cells and platelet-rich plasma combination for the treatment of supraspinatus tendinopathy in 55 dogs: a retrospective study. Front Vet Sci 2016;3:61.
2. Carr BJ, Canapp SO, Canapp DA, et al. Adhesive capsulitis in eight dogs: diagnosis and management. Front Vet Sci 2016;3:55.
3. Leeman JJ, Shaw KK, Mison MB, et al. Extracorporeal shockwave therapy and therapeutic exercise for supraspinatus and biceps tendinopathies in 29 dogs. Vet Rec 2016;179:385.
4. Spall BF, Fransson BA, Martinez SA, et al. Tendon volume determination on magnetic resonance imaging of supraspinatus tendinopathy. Vet Surg 2016;45:386–391.
5. Ho LK, Baltzer WI, Nemanic S, et al. Single ultrasound-guided platelet-rich plasma injection for treatment of supraspinatus tendinopathy in dogs. Can Vet J 2015;56:845–849.
6. Mathewson MA, Kwan A, Eng CM, et al. Comparison of rotator cuff muscle architecture between humans and other selected vertebrate species. J Exp Biol 2014;217:261–273.
7. Maddox TW, May C, Keeley BJ, et al. Comparison between shoulder computed tomography and clinical findings in 89 dogs presented for thoracic limb lameness. Vet Radiol Ultrasound 2013;54:358–364.
8. Mistieri ML, Wigger A, Canola JC, et al. Ultrasonographic evaluation of canine supraspinatus calcifying tendinosis. J Am Anim Hosp Assoc 2012;48:405–410.
9. Lafuente MP, Fransson BA, Lincoln JD, et al. Surgical treatment of mineralized and nonmineralized supraspinatus tendinopathy in twenty-four dogs. Vet Surg 2009;38:380–387.
10. Murphy SE, Ballegeer EA, Forrest LJ, et al. Magnetic resonance imaging findings in dogs with confirmed shoulder pathology. Vet Surg 2008;37:631–638.
11. Fransson BA, Gavin PR, Lahmers KK. Supraspinatus tendinosis associated with biceps brachii tendon displacement in a dog. J Am Vet Med Assoc 2005;227:1429–1433.
12. Sager M, Herten M, Ruchay S, et al. The anatomy of the glenoid labrum: a comparison between human and dog. Comp Med 2009;59:465–475.
13. Gombera MM, Sekiya JK. Rotator cuff tear and glenohumeral instability: a systematic review (Erratum published in Clin Orthop Relat Res 2015;473:751). Clin Orthop Relat Res 2014;472:2448–2456.
14. Goldberg JA, Chan KY, Best JP, et al. Surgical management of large rotator cuff tears combined with instability in elite rugby football players. Br J Sports Med 2003;37:179–181.
15. Porcellini G, Caranzano F, Campi F, et al. Glenohumeral instability and rotator cuff tear. Sports Med Arthrosc Rev 2011;19:395–400.
16. Toolanen G, Hildingsson C, Hedlund T, et al. Early complications after anterior dislocation of the shoulder in patients over 40 years: an ultrasonographic and electromyographic study. Acta Orthop Scand 1993;64:549–552.
17. Travlos J, Goldberg I, Boome RS. Brachial plexus lesions associated with dislocated shoulders. J Bone Joint Surg Br 1990;72:68–71.
18. Baumgarten KM, Gerlach D, Galatz LM, et al. Cigarette smoking increases the risk for rotator cuff tears. Clin Orthop Relat Res 2010;468:1534–1541.
19. Linaker CH, Walker-Bone K. Shoulder disorders and occupation. Best Pract Res Clin Rheumatol 2015;29:405–423.
20. Deutsch A, Altchek DW, Veltri DM, et al. Traumatic tears of the subscapularis tendon. Clinical diagnosis, magnetic resonance imaging findings, and operative treatment. Am J Sports Med 1997;25:13–22.
21. Potter HG, Birchansky SB. Magnetic imaging of the shoulder: a tailored approach. Tech Shoulder Elbow Surg 2006;6:43–56.
22. Goutallier D, Postel JM, Bernageau J, et al. Fatty muscle degeneration in cuff ruptures. Pre- and postoperative evaluation by CT scan. Clin Orthop Relat Res 1994;304:78–83.
23. Evans HE, de Lahunta A. The muscular system. In: Miller's anatomy of the dog. 4th ed. St Louis: Elsevier Health Science, 2013:235.
24. Gillard GC, Reilly HC, Bell-Booth PG, et al. The influence of mechanical forces on the glycosaminoglycan content of the rabbit flexor digitorum profundus tendon. Connect Tissue Res 1979;7:37–46.
25. Fan L, Sarkar K, Franks DJ, et al. Estimation of total collagen and types I and III collagen in canine rotator cuff tendons. Calcif Tissue Int 1997;61:223–229.
26. Apostolakos J, Durant TJ, Dwyer CR, et al. The enthesis: a review of the tendon-to-bone insertion. Muscles Ligaments Tendons J 2014;4:333–342.
27. Peh WC, Chan JH. The magic angle phenomenon in tendons: effect of varying the MR echo time. Br J Radiol 1998;71:31–36.
28. Maffulli N, Barrass V, Ewen SW. Light microscopic histology of achilles tendon ruptures. A comparison with unruptured tendons. Am J Sports Med 2000;28:857–863.
29. Laitinen OM, Flo GL. Mineralization of the supraspinatus tendon in dogs: a long-term follow-up. J Am Anim Hosp Assoc 2000;36:262–267.
30. Oliva F, Via AG, Maffulli N. Physiopathology of intratendinous calcific deposition. BMC Med 2012;10:95.
31. Frannson BA. Supraspinatus tendinosis and mineralization: same disease two faces? In: Proceedings. American College of Veterinary Surgeons Veterinary Symposium, the Surgical Summit 2011;362–364.
32. Nishimura T, Hattori A, Takahashi K. Relationship between degradation of proteoglycans and weakening of the intramuscular connective tissue during post-mortem ageing of beef. Meat Sci 1996;42:251–260.
33. Webbon PM. A post mortem study of equine digital flexor tendons. Equine Vet J 1977;9:61–67.
Appendix
Parameters used for various MRI fluid-sensitive pulse sequences obtained of both shoulder joints of 7 nonlame 2-year-old sexually intact male Beagle cadavers.
| MRI sequence | ||||||
|---|---|---|---|---|---|---|
| Parameter | Proton-density FSE | T2-weighted spectral selective fat suppression | T2-weighted | Short tau inversion recovery | Multiplanar gradient recall acquisition in the steady state | Spoiled gradient recalled acquisition |
| Echo time (ms) | 20 | 37 | 84 | 19 | 20 | 2.5 |
| Repetition time (ms) | 1,000–4,000 | 5,500 | 1,000–3,000 | 4,176 | 600 | 11.9 |
| Field of view frequency (cm) | 10 | 10 | 10 | 10 | 10 | 8–10 |
| Field of view phase (cm) | 10 | 10 | 10 | 10 | 10 | 8–10 |
| Flip angle (°) | 111 | 111 | 111 | 111 | 45 | 20 |
| Inversion time (ms) | — | — | — | 190 | — | — |
| Matrix (frequency encoding) | 512–416 | 288 | 512 | 288 | 512 | 384 |
| Matrix (phase encoding) | 384 | 288 | 384 | 288 | 224 | 384 |
| No. of excitations | 1.5 | 1 | 1.5 | 1 | 1 | 1 |
| Echo train length (No. of echoes) | 16 | 16 | 15 | 10 | 1 | 1 |
| Bandwidth (Hz/pixel) | 325 | 162 | 325 | 162 | 75 | 278 |
| Slice thickness (mm) | 1.5 | 2 | 1.5 | 1.5 | 1.5 | 0.6 |
All cadavers were stored at 4°C and scanned within 72 hours after euthanasia. — = Not applicable.
