Clinique vétérinaire

du Dr Bardet

Diagnosis of Shoulder Instability in Dogs and Cats: A Retrospective Study

The glenohumeral joint is a remarkable articulation providing the greatest range of motion of any joint in the body. Glenohumeral stability results from several mechanisms, including those that do not require expenditure of energy by muscle ("passive mechanisms") and those that do ("active mechanisms"). Glenohumeral instability has been recognized in 47 shoulders of 45 dogs and one cat. Cases are presented because of chronic foreleg lameness. Shoulder joint pain is obviated by the orthopedic examination. Only 57% of the involved shoulders presented with degenerative joint disease. Signs of instability are recognized under anesthesia using a craniocaudal or mediolateral drawer sign or both. This report describes the radiographic and arthroscopic findings of shoulder instability. Arthroscopy of the shoulder joint allows identification of all intra-articular pathologies. Shoulder instability, not fully recognized in the past, appears to be the most common cause of shoulder lameness in the dog. J Am Anim Hosp Assoc 1998;34:42-54.

Introduction

The shoulder joint is a common source of foreleg lameness. When shoulder pain is elicited, many cases of foreleg lameness end up in a diagnostic "blind alley (1). It is very difficult to make an exact diagnosis due to the lack of objective findings to reinforce the subjective signs (1). Many cases are treated successfully based on a presumptive diagnosis of tenosynovitis of the biceps tendon (1). Surprisingly, the veterinary literature focused primarily on treatment of shoulder luxations (1-19), with only a few papers on the anatomy and biomechanics of the canine shoulder joint (9,20-22). Descriptions of shoulder subluxations are anecdotal (23-27).

In 1986, the first publication on arthroscopy of the canine shoulder gave details of the technique, normal anatomy, and complications (28). Several later papers have described the use of diagnostic arthroscopy of the shoulder and arthroscopic surgery for the treatment of osteochondritis dissecans (OCD) of the shoulder joint in dogs (29-35).

In human orthopedics, few topics have as broad an outline as the treatment of the unstable shoulder. This subject has produced decades of debate and over 250 surgical techniques (36-39). The past decade has produced many advances in the diagnosis and treatment of shoulder instabilities including better anatomical studies, (40) which describe the anatomy of stability as well as the pathophysiology of instability; the use of magnetic resonance (MR) imaging to describe the multitude of bony and soft-tissue lesions found in the unstable shoulder (41), the refinement of the medical history and the physical examination to provide a more efficient diagnosis; and the use of the arthroscope to finalize the pathoanatomic diagnosis and to treat the unstable shoulder (12-45). The evolution of open surgical techniques and postoperative physiotherapy leads to a higher quality outcome on a predictable basis.

The goals of this paper are to review the surgical anatomy and biomechanics related to shoulder stability and to describe the diagnosis of shoulder instability, focusing mainly on arthroscopic signs of instability.

Anatomy and Biomechanics of Glenohumeral Stability

The glenohumeral joint is suited for mobility. The large, spherical head of the humerus articulates with the small, shallow glenoid fossa of the scapula. The glenoid provides little coverage of the humeral head (46). It has been suggested that glenohumeral stability results from a hierarchy of mechanisms, including those that do not require the expenditure of energy by muscle ("passive mechanisms") and those that do ("active mechanisms") (40). The shoulder joint is capable of movement in any direction, but its chief movements are flexion and extension (20,46).

Passive Mechanisms

Muscle activity is not required to hold the shoulder together. The intact shoulder of a fresh, anatomical specimen is quite stable. It appears appropriate to discuss the "passive" mechanisms of the glenohumeral joint which include ligamentous and capsular restraints, joint conformity, glenoid labrum, joint conformity, finite joint volume, and adhesion/cohesion.

The glenohumeral ligaments can be identified on the deep surface of the articular capsule on the medial and lateral sides of the shoulder joint (9,46). The medial glenohumeral ligament (MGHL) appears either as a Y shape [Figure 1] or a transverse band [Figures 2, 3]. The MGHL extends downward from the medial sur-face of the supraglenoid tubercle of the scapula [Figure 3] across the shoulder joint to attach to the joint capsule at the junction of the humeral neck and lesser tubercule (9). The caudal band of the ligament attaches proximally (approximately 2 cm caudal to the cranial band) on the medial side of the glenoid cavity. The MGHL appears more often as a Y shape in large-breed dogs. On arthroscopy from a lateral portal, the MGHL appears most commonly as a transverse diagonal band on the medial side of the glenohumeral joint [Figure 4]. The MGHL is visualized best when the joint is distended. The distal humeral insertion is located caudally in relation to the tendon of insertion of the subscapularis muscle [Figure 5].

Figure 1 - Macroscopic view of an anatomical specimen of the left shoulder joint showing the medial glenohumeral ligament (MGHL). Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, (3) Y-shaped MGHL and (4) supraglenoid tubercle.
Figure 2 - Macroscopic view of an anatomical specimen of the left shoulder joint. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, (3) medial glenohumeral ligament (MGHL) which appears as a diagonal band on the medial aspect of the joint, (4) biceps tendon, and (5) tendon of insertion of the subscapularis muscle.
Figure 3 - Detailed macroscopic view of the craniomedial aspect of the left shoulder. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, (3) medial glenohumeral ligament (MGHL) which appears as a diagonal band on the medial aspect of the joint, (4) biceps tendon, (5) tendon of insertion of the subscapularis muscle, and (6) supraglenoid tubercle.
Figure 4 - Arthroscopic view of the medial aspect of the left shoulder joint from a lateral portal. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, and (3) medial glenohumeral ligament (MGHL) which appears as a diagonal band on the medial aspect of the joint.
Figure 5 - Arthroscopic view of the craniomedial aspect of the left shoulder joint from a lateral portal. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, (3) medial glenohumeral ligament (MGHL) which appears as a diagonal band on the medial aspect of the joint, (4) biceps tendon, and (5) tendon of insertion of the subscapularis muscle.

The lateral glenohumeral ligament (LGHL) extends downward from the lateral rim of the glenoid cavity to attach to the neck of the humerus and to the caudal portion of the greater tubercle [Figure 6]. In large-breed dogs, the ligament is approximately 2 cm wide at its proximal insertion and 1.5 cm wide at its humeral attachment (9). Seen from either a medial or a caudolateral portal on arthroscopy, the LGHL appears as a wide ligament on the craniolateral aspect of the glenohumeral joint [Figure 7].

Figure 6 - Macroscopic view from an anatomical specimen of the craniolateral aspect of the shoulder. Note the (1) articular cartilage of the humeral head, (2) biceps tendon, (3) bicipital groove, (4) lateral glenohumeral ligament (LGHL), and (5) joint capsule.
Figure 7 - Arthroscopic view of the lateral glenohumeral ligament (LGHL). Note the (1) LGHL and (2) articular cartilage of the humeral head.

The medial and lateral glenohumeral ligaments appear. as relatively wide structures with strands of fibers that invaginate in the joint cavity from the capsule so that the inner and outer surfaces of the ligaments are covered with synovial membranes (9).

The articular joint capsule forms a loose sleeve which attaches proximally at the periphery of the glenoid cavity on the cranial, lateral, and caudal aspects [Figure 8]. On the medial margin, the joint capsule attaches more proximally several millimeters away from the glenoid rim, forming a synovial recess [Figures 9, 10]; on the humeral neck, it attaches several millimeters distal to the articular aspect of the humeral head, where it blends with the periosteum on the neck of the humerus. Part of the joint capsule surrounds the tendon of origin of the biceps brachii muscle and extends distally about 2 cm in the intertubercular groove [Figures 3, 11]. Medially, the capsule is attached loosely to the tendons of the subscapularis and coracobrachialis muscles and laterally to the tendons of the infraspinatus and teres minor muscles. These tendons can be separated carefully from the capsule without entering the joint.

Figure 8 - Macroscopic view from an anatomical specimen of the glenoid cavity and capsuloligamentous restraints of the left shoulder. Note the (1) glenoid cavity, (2) biceps tendon, (3) medial glenohumeral ligament (MGHL), (4) caudomedial joint capsule, and (5) lateral joint capsule.
Figure 9 - Macroscopic view from an anatomical specimen of the glenoid cavity and capsuloligamentous restraints of the left shoulder as seen in Figure 8. Note the (1) glenoid cavity, (2) biceps tendon, (3) medial glenohumeral ligament (MGHL), (4) caudomedial joint capsule, (5) lateral joint capsule, and (6) scalpel handle introduced into the capsular recess.
Figure 10 - Macroscopic view from an anatomical specimen of the medial aspect of the left glenoid cavity. Note the (1) glenoid cavity, (2) biceps tendon, (3) medial glenohumeral ligament (MGHL), (4) caudomedial joint capsule, and (5) medial free margin of the glenoid cavity. The joint capsule inserts on the scapular neck.
Figure 11 - Arthroscopic view of the biceps tendon and bicipital groove of the right shoulder joint. Note the (1) cranial surface of the articular cartilage of the humeral head, (2) biceps tendon, and (3) bicipital groove.

Until recently, it was agreed that the insertions of the "cuff muscles" were responsible for maintaining joint integrity (23). More recently it has been shown that the joint capsule and collateral ligaments of the glenohumeral joints play significant roles in stability (9,21). An in vitro study revealed that cutting the tendons of the "cuff muscles" results in minimal loss of stability. Stability is decreased substantially by transecting the capsule and collateral ligaments (21). The humeral joint is not luxated easily when the collateral ligaments are intact (21).

Other contributing mechanisms to stability are con-cavity and compression (47). The depth of the bony glenoid is enhanced by the articular cartilage and the glenoid labrum when present. The concavity and the fit of the glenoid to the humeral head provide stability to the joint, which is enhanced by forces pressing the ball into the socket (47) (see Active Mechanisms).

The glenoid labrum is a fibrous rim that serves to deepen the glenoid fossa and allow attachment of the glenohumeral ligaments and the biceps tendon to the glenoid in humans (47). It is the interconnection of the glenoid periosteum, bone, articular cartilage, synovium, and joint capsule. This labrum is described in dogs (9,23,46); however, the authors only identified it macroscopically in one of the 12 cadaver specimens used for the anatomical study of the shoulder joint [Figure 12]. The labrum is described as fibrocartilage that extends 1 to 2 mm. beyond the edge of the glenoid cavity caudolaterally (46). It may appear as the proximal insertion of the biceps tendon [Figure 13], but in most cases there is no macroscopic transitional structure at the insertion of the joint capsule on the glenoid rim [Figure 14]. Therefore, the stabilizing role of the labrum in the dog appears limited.

Figure 12 - Macroscopic view from an anatomical specimen of the right glenoid cavity and periarticular labrum. Note the (1) glenoid cavity, (2) biceps tendon, (3) medial capsular recess, (4) caudolateral labrum, and (5) caudolateral joint capsule.
Figure 13 - Arthroscopic view of the biceps tendon of the left shoulder. Note the (1) supraglenoid tubercle, (2) cranial border of the articular cartilage of the humeral head, (3) biceps tendon, and (4) glenoid labrum.
Figure 14 - Arthroscopic view of the caudal margin of the glenoid cavity. Note the (1) articular cartilage of the glenoid cavity, (2) joint capsule, and the absence of an obvious labrum.

Anatomical studies, surgical findings, attempts at aspiration, and MR images confirm that there is mini-mal (less than 1 ml) free fluid in the normal shoulder joint (47). The normal shoulder is sealed by the capsule. Thus, like the syringe, the shoulder joint is stabilized by its limited joint volume. As long as the joint is a closed space containing minimal free fluid, the joint surfaces cannot be distracted easily or subluxated. Small translations of the humerus on the glenoid can be balanced by fluid flow in the opposite direction, allowing a nonuniform gap to open in the joint space. This gap can increase until all available fluid has been mobilized, at which point further motion of the joint is resisted by negative fluid pressure in the joint. This negative pressure pulls the capsule inward to-ward the joint space, putting its fibers "on the stretch." This mechanism is aided by the fact that intraarticular pressure normally is slightly negative (49).

The stabilization mechanism changes when the gap between the articular surfaces becomes very small. Viscous and intermolecular forces begin to dominate, preventing ready fluid motion and providing a cohesive bond between the glenoid and humerus. This is termed the "adhesion-cohesion mechanism" (47). A familiar example is provided by two wet microscope slides pressed together. Water is held to their surfaces by adhesion. They readily can slide on each other but cannot be pulled apart easily by forces applied at right angles to their flat surfaces, because the water holds them together by cohesion. Joint surfaces are wet with joint fluid that holds them together by adhesion/cohesion as well. This joint fluid interface has the very desirable properties of high tensile strength (i.e., surfaces are difficult to pull apart), and little shear strength (i.e., the interface allows sliding of the two joint surfaces on each other with low resistance). An important distinction is that the adhesion/cohesion mechanism does not put the capsular fibers on the stretch, as viscous forces suffice to prevent fluid from entering the joint space. Thus, stability is pro-vided entirely by forces exerted by and on the articular surfaces.

Both the effects of limited joint volume and adhesion/cohesion would be reduced or eliminated by the addition of excess fluid (gas or liquid) to the joint or when the joint is inflamed. This phenomenon was well described by Humphry in 1858 (47).

Active Mechanisms

Dynamic glenohumeral stability in humans is provided by the biceps and the subscapularis, supraspinatus, infraspinatus, and teres minor muscles of the rotator cuff. The cuff muscles serve several stabilizing functions. First, by virtue of the blending, of their tendons with the glenohumeral capsule and ligaments, selective contraction of the cuff muscles can adjust the tension in these structures, producing "dynamic" ligaments, as proposed by Cleland in 1866 (47) Second, by contracting together, they press the humeral head into the glenoid fossa, locking it into position and thus providing a secure scapulohumeral link for fore-limb function. Third, by contracting selectively, the rotator cuff muscles can resist displacing forces resulting from contraction of the principal shoulder muscles.

Materials and Methods

Forty-seven shoulder joints were examined in 45 dogs and one cat between September 1, 1993 and March 31, 1996 because of foreleg lameness and shoulder instability. The history included the breed, age, sex, weight, onset and duration of clinical signs, progression of signs, activity of the animal, prior medical treatment, and influence of activity. Ages of dogs ranged from 18 months to 13 years (mean, 5.3 years). There were 30 males and 16 females. Among the 45 dogs, 23 breeds were represented including Brittany (n=6), French poodle (n=5), Labrador retriever (n=5), and Doberman pinscher (n=4) [Table 1]. Only one dog (2.2%) had bilateral involvement. Before entering the study, each case underwent complete physical and orthopedic examinations. In two patients, clinical signs initially were attributed to cervical disk disease. A myelographic examination was performed first, followed by an electromyography and nerve conduction velocities.

Table 1

When the cause of lameness was located, the orthopedic examination included range of motion, presence of pain in hyperextension, biceps tendon test, and palpation to assess joint instability. Each animal was anesthetized, and the same orthopedic examination was repeated. Mediolateral, craniocaudal, and stress mediolateral radiographs were taken of each shoulder. The preoperative radiographic status of osteoarthritis involving the shoulder joint was graded as absent, minimal (i.e., osteophytes less than 2 mm), moderate (i.e., osteophytes 2-to-4 mm), or severe (i.e., osteophytes greater than 4 mm). Following radiography, an arthroscopy of the affected joint was performed prior to any treatment.

Arthroscopic examination was performed with a 2.7-mm 30° foreoblique arthroscope with a 3.5-mm, outside-diameter sleeve. Light was supplied by a xenon source. The arthroscopic procedure was visualized on a monitor using a camera. Photographic documentation was made with a color printer. Each patient was positioned in lateral recumbency and the leg was prepared aseptically. The joint was distended with 10-to-15 ml of lactated Ringer's solution after being punctured with a 19-gauge needle craniolaterally between the acromion and caudal part of the greater tubercule in a caudomedial direction. A stab incision was made 1 cm caudally and I cm distally to the acromion (28) using a no. 11 Bard-Parker scalpel blade. The joint capsule then was penetrated using the blunt trocar locked in the arthroscopic sleeve. The trocar was replaced by the arthroscope, and the light cable and the camera and inflow lines were connected. Joint inspection then could be performed. Fluid inflow was maintained via a sterile infusion set connected to the stopcock of the trocar sleeve. Fluid leaving the outflow canula was drained away via a silastic tube (21).

With the arthroscope in the lateral portal, a systematic arthroscopic examination is carried out with inspection of the synovium, articular cartilage surfaces, glenohumeral ligaments, labrum, tendon of the biceps muscle, tendon of the subscapularis muscle, and joint capsule.

Results

Each dog and the cat were presented because of chronic foreleg lameness. Most had been lame for more than two months and some for several years. The lameness was permanent (i.e., continual and nonresponsive to nonsteroidal anti-inflammatory drug [NSAID] treatment) in 35 dogs and one cat and intermittent in 10 dogs. Five dogs and the cat had nonweight-bearing lamenesses. Five dogs were crying spontaneously because of pain and were referred for cervical disk disease; two of these five were referred for tetraplegia. One of these five, a Brittany, presented for evaluation of cervical disk disease, it was walking as if it had wobbler syndrome, was unable to go downstairs, and was jumping from 1.5 m-high stairs and landing on all four legs. This case had bilateral shoulder involvement. Atrophy of the shoulder muscles was obvious in 15 dogs. The clinical signs are summarized in Table 2.

Table 2

All dogs except one had pain on shoulder hyperextension. The biceps tendon test, with the front limb in full extension, along with the thorax test were positive in 40 shoulders. In five shoulders, the subluxation was recognized without anesthesia during the biceps tendon test.

A preoperative radiographic examination was performed for each shoulder. Normal shoulders were identified in 20 (43%) cases [Table 3]. Osteoarthritis was observed in 27 (57%) cases. Osteoarthritis was classified as minimal in 10 (21%) cases and severe in 12 (26%) cases. A medial osseous defect was observed on the craniocaudal views of the humeral head in five (11%) dogs, and calcification of the tendon of the supraspinatus muscle was obvious in five (11 %) more cases. The medial rim of the glenoid cavity appeared flattened in three cases.

Table 3

Upon arthroscopic examination, the synovial membrane was normal in 15 of the 47 shoulders [Table 4]. In the remaining 32 shoulders, synovitis was graded as minimal (n=10), moderate (n=5), severe (n=13) [Figure 15], and fibrous (n=4). Abnormalities of the articular cartilage were observed on the glenoid cavity and on the humeral head. The most common finding on the glenoid cavity was an erosion of the medial rim (n=22) [Figure 16]. The articular cartilage of the glenoid cavity was eburnated (n=8); eburnation always was associated with osteophyte formation on the caudal rim of the glenoid cavity. The most common articular cartilage defect was located on the caudal articular cartilage of the humeral head in 25 shoulders [Figure 17]. Eburnation of the humeral head was associated with osteophyte formation in 11 shoulders. In three patients, the osseous defect seen on the craniocaudal radiographs was observed during the arthroscopic examination.

Table 4
Figure 15 - Arthroscopic view of the left shoulder showing a severe synovitis associated with shoulder instability. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, (3) medial glenohumeral ligament (MGHL) which appears as a diagonal band on the medial aspect of the shoulder, and (4) severe synovitis with pendulous villi.
Figure 16 - Arthroscopic view of the left shoulder showing erosions of the medial margin of the glenoid cavity. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, (3) medial glenohumeral ligament (MGHL) which appears as a diagonal band on the medial aspect of the shoulder, and (4) severe synovitis with pendulous villi.
Figure 17 - Arthroscopic view of the caudal aspect of the right shoulder showing erosion of the caudal articular surface of the humeral head. Note the (1) caudal margin of the glenoid cavity, (2) caudal articular surface of the humeral head, and (3) joint capsule.

After assessing the synovium and articular cartilage, the evaluation of the glenohumeral ligaments was carried out. The medial glenohumeral ligament (MGHL) either was distended [Figures 18A, 18B] and frayed (n=20) [Figures 19, 20] or torn (n=8) [Figures 21, 22]. In three cases, only the caudal part of the MGHL was thickened [Figure 23]. If no lesion was identified in the MGHL, the lateral and caudolateral labrum was inspected. It was found detached in five shoulders [Figure 24].

After assessing the biceps anchor, the biceps is followed out to its exit point in the bicipital groove. The biceps tendon was torn partially in three cases and bipartite in one case. In one shoulder, a severe tendonitis without rupture was observed. The tendon of insertion of the subscapularis muscle was torn in two cases [Figure 25]; a tendonitis of the same tendon was seen in three instances.

The arthroscopist should be familiar with capsular volume. The caudal axillary recess was capacious and did not tighten with appropriate changes in leg position, suggesting the possibility of a plastically de-formed capsule secondary to microtrauma [Table 4] in three shoulders.

Figure 18A
Figures 18A, 18B - Arthroscopic view of the right shoulder showing a distended, incompetent medial glenohumeral ligament (MGHL). (A) Note the (1) glenoid cavity, (2) humeral head, (3) distended MGHL, and (4) joint capsule. (B) Note the (1) glenoid cavity, (2) humeral head, (3) distended MGHL, and (4) an egress cannula probing the incompetent MGHL.
Figure 19 - Arthroscopic view of a frayed medial glenohumeral ligament (MGHL) in the left shoulder. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, and (3) MGHL. The MGHL is torn partially.
Figure 20 - Arthroscopic view of a frayed medial glenohumeral ligament (MGHL) in the right shoulder. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head. (3) MGHL, and (4) an egress cannula probing the dilacerated MGHL.
Figure 21 - Arthroscopic view of a torn medial glenohumeral ligament (MGHL) in the right shoulder. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, and (3) MGHL.
Figure 22 - Arthroscopic view of the medial aspect of the left shoulder. The medial glenohumeral ligament (MGHL) is absent. Note the (1) articular cartilage of the glenoid cavity, (2) articular cartilage of the humeral head, (3) medial joint capsule, and (4) an egress cannula probing the distended joint capsule.
Figure 23 - Arthroscopic view of the medial aspect of the right shoulder showing the thickening of the caudal branch of the medial glenohumeral ligament (MGHL). Note the (1) glenoid cavity, (2) humeral head, and (3) thickened caudal branch of the MGHL.
Figure 24 - Arthroscopic view of the detached labrum and joint capsule from the lateral aspect of the right shoulder. Note the (1) lateral margin of the glenoid cavity, (2) humeral head, (3) detached labrum and joint capsule, and (4) an egress cannula probing the lateral joint capsule tear.
Figure 25 - Arthroscopic view of the craniomedial aspect of the left shoulder joint showing the torn insertion of the subscapularis muscle tendon. Note the (1) biceps tendon, (2) humeral head, (3) torn subscapularis muscle tendon, and (4) joint capsule.

Discussion

Shoulder instability appears to be a common cause of lameness in dogs, affecting mainly medium- and large-breed dogs. Most affected dogs are hyperactive. The Brittany, French poodle, and Labrador retriever appear overrepresented. Shoulder instability is much more common than luxations since only nine dogs were treated for shoulder luxations during the same period. The most common clinical presentation is a permanent foreleg lameness. It should be differentiated from spinal disorders since five cases were referred because of neurological disorders.

Pain on hyperextension was present in almost every case, and the biceps tendon test was positive in 40 (85%) of 47 unstable shoulders. The biceps tendon test appears to be more an indicator of shoulder joint pain than a pathognomonic sign of biceps tendon disorders. In five shoulders, subluxations were recognized preoperatively without anesthesia. Having learned to diagnose shoulder instability initially by arthroscopy, a clinical test of palpation under anesthesia was developed to detect shoulder subluxations. It may be compared with the drawer sign for the stifle joint. Both shoulders should be compared in the same individual. It is mandatory to use bony landmarks of the scapula and humerus. The scapula is held in one hand with the thumb on the acromion and the index finger wrapped around the craniomedial side of the scapular neck. The other hand holds the humerus with the thumb on the caudolateral aspect of the proximal humeral metaphysis and the index finger on the greater tubercule. Translocation of the humeral head in the cranial, medial, caudal, and lateral directions is determined with the shoulder in a semiflexed position. A craniocaudal or mediolateral drawer sign may be elicited. The direction and importance of the sub-luxation must be recorded. The degree of translocation may be classified as none or absent (i.e., Grade 1) when translocation of the head of the humerus on the glenohumeral joint is not appreciated; as mild (i.e., Grade 2) when some translocation is appreciated but is not enough to allow the head of the humerus to rise up on the rim of the glenoid cavity; as moderate (i.e., Grade 3) when the head of the humerus is appreciated to move up on the rim of the glenoid cavity; and as severe (i.e., Grade 4) when the head of the humerus courses over the rim of the glenoid cavity and is dislocated (50).

Degenerative joint disease (DJD) was present in 57% of the cases. Degenerative joint disease was minimal in 21% of the unstable shoulders; the first sign was a small osteophyte on the caudal margin of the humeral head on the mediolateral radiograph. When DJD is apparent on the radiographs and occurs in the absence of OCD, instability of the shoulder joint should be suspected strongly. However, the absence of DJD cannot exclude instability since 43% of the unstable shoulders did not show any signs of DJD. Other means of diagnosis are most helpful for the early diagnosis of shoulder instability.

Shoulder instability in humans typically is diagnosed and classified on the basis of history, physical examination, and plain radiographs. Classification of the instability is of considerable importance because surgical treatment varies with the direction of the instability. Special diagnostic tests are available in human medicine. They include stress examination, tomographic or computed tomographic arthrography, ultrasonography, MR imaging, shoulder arthroscopy, and exploratory arthrotomy (41,51). Plain film radiographs primarily assess the alignment and integrity of the osseous structures. Ultrasonography has been advocated for the detection of rotator cuff tears, whereas computed tomography (CT) after the injection of intraarticular contrast has been used in the detection of injuries to the osseous or labral-capsular components (50). The most recent modality to be employed is MR imaging, a technique that combines excellent soft-tissue contrast and multiplanar capabilities without the use of ionizing radiation (41,50). However, these techniques are not readily available in most veterinary hospitals, or the techniques are not described. None of the tests are effective for assessing abnormalities of the joint capsule, nor can they assess the competency of the capsule and its ligamentous structures in preventing translocation of the head of the humerus on the glenoid cavity. Shoulder arthroscopy appears effective for identifying impression lesions of the head of the humerus and glenoid margins and for assessing the volume and relative laxity of the capsule, anomalies of the synovium, and lesions of the glenohumeral ligament, biceps tendons, and labrum.

The glenohumeral joint is a remarkable articulation, providing the greatest range of motion of any joint in the body. In contrast to the acetabulum of the hip joint, the glenoid does not provide an instrinsically stable socket. Capsuloligamentous constraints provide a critical contribution to glenohumeral joint stability (21,37,51). The concept of concavity compression refers to the stability afforded a convex object that is pressed into a concave surface. For example, pressing a round ball onto a flat table top provides little resistance to a counter pressure trying to slide the ball across the table's surface. If the ball is pressed into a concavity on the table top, the stability of the ball is enhanced by the depth of the concavity and by the magnitude of the compressive force. The glenoid articular geometry may seem too shallow to provide significant constraint for the humeral head. Although the glenoid fossa is only one-fourth the size of the humeral articular surface, it does provide a concavity (51). Stability attributed to concavity compression is compromised if the glenoid is small or flat, if the labrum is torn or avulsed, or when the concavity is lessened by injury or wear as observed radiographically in three cases. Recurrent instability episodes would tend to erode the articular cartilage of the medial glenoid rim and further lessen the concavity. Perhaps relative flatness of the glenoid articular sur-face (i.e., lack of effective glenoid depth) could predispose these patients to abnormal subluxations.

Simple observation suggests that the harder the humeral head is compressed into the glenoid concavity, the more stable the glenohumeral joint is to applied translating forces. The various shoulder muscles provide the dynamic compression of the humeral head into the glenoid concavity in vivo. Other potential causes of subluxations related to an abnormal concavity compression include muscle imbalance, creating a net force that deviates excessively from the glenoid center, or abnormal position of the humeral head due glenoid version.

High-demand, repetitive use of the shoulder also has been implicated in the etiology of glenohumeral instability. Historically, shoulder dislocations had been classified as either "traumatic" or "atraumatic". Shoulders were thought to become unstable either on the basis of a major traumatic injury or multiple small traumas. Repetitive microtrauma has been implicated as an etiology of shoulder instability, mainly in athletes who use their arms to throw objects (52). If these stresses are applied at a rate that is greater than the rate of tissue repair, these repetitive insults can produce damage to the tissues. It is well accepted that if a material is subjected to a large number of loading cycles, it will fail at a stress lower than its ultimate tensile stress. It is quite possible that the repetitive, high-velocity motions of the shoulder may cause fatigue failure to the fibers of the glenohumeral ligaments because the endurance limit is exceeded during these motions (52). It has been shown that a ligament undergoes significant stretching before ultimate failure when it is tested in uniaxial tension (52). It is suspected that the high-demand, repetitive loading of certain shoulders may lead to fatigue failure of the ligament, resulting in stretching of the ligament and impairment of the proprioceptive function of the capsule. Indeed, axonal fibers of different diameters have been identified in the glenohumeral ligaments, suggesting a proprioceptive role for these ligaments (53). Moreover, differences have been demonstrated in shoulder proprioception between stable shoulders and unstable shoulders before and after repair (52).

In 31 (66%) of 47 cases, the MGHL either was distended (i.e., incompetent), torn, or thickened; in five (11%) cases, there was a tear of the caudolateral labrum. The severity of the MGHL lesions appeared to be related to the degree of DJD. When the MGHL was torn, in two instances the tendon of insertion of the subscapularis muscle also was torn, the shoulder joint was severely osteoarthritic with eburnation and an associated fibrous synovitis. Changes related to aging in soft tissues of the shoulder may represent a complex situation in which two independent factors (i.e., chronological aging and activity-induced wear-and-tear effects) might contribute in equally significant ways to the development of the pathologies (54). Aging is associated with important structural, bio-chemical, and biomechanical changes in the tendon (55). The collagen content of a tendon remains fairly constant throughout maturity and seems to decrease slowly with age (53). In old rabbits, multiple morphological and biochemical changes occur in the Achilles tendon which are attributed to aging (54). In older humans, the supraspinatus tendon shows a progressive loss in integrity with marked fiber disorganization (54). Aging may be a contributing factor of shoulder instability in dogs; however, not every old dog has un-stable shoulders even if many have DJD of the shoulder without clinical signs (56). Many younger dogs (mean age, 5.3 years) have unstable shoulders.

It also has been shown that dogs suffering from hip dysplasia may suffer from shoulder DJD three-to-four years after the clinical manifestations of hip dysplasia (56). Several joints appear involved by DJD. Genetic factors may be suspected as in hip dysplasia.

Joint instability has been recognized as a cause of DJD. The phenomenon of joint laxity and instability, leading to altered contact stresses and contact areas, has been well documented (56,57). The exact role of repetitive trauma on articular cartilage has been questioned. Repetitive loading appears to be a factor in the development of the osteoarthritis of the glenohumeral joint (52,58). Many shoulders are subject to repetitive stresses, but only a small percentage of cases develop clinical signs.

Finally, the instability of the shoulder joint should be differentiated from the tenosynovitis of the biceps tendon,' mineralization of the supraspinatus tendon (59,60), calcifying tendinopathy of the biceps brachii muscle (61), and rupture of the biceps brachii tendon (1). Tenosynovitis of the biceps tendon was a disease of the 1940s and 1950s in the human medical literature (62). The number of diagnosed cases has decreased dramatically over the decades because of the improvement in the knowledge of the anatomy, biomechanics, and diagnostic imaging of the shoulder joint. During this study, the author found only one case of tenosynovitis of the biceps tendon (63). Arthroscopy offers the advantage of recognizing all the intraarticular pathologies. Most of these pathologies are recognized today because of the magnification offered by new arthroscopic means and were not recognized in the past, even after exploratory arthrotomy of the shoulder joint. Mineralization of the supraspinatous tendon and calcifying tendinopathy may be suspected if all the other intraarticular causes of lameness have been eliminated by arthroscopy.

Conclusion

Shoulder instability was found to be the most common cause of shoulder lameness in medium- and large-breed adult dogs. The patients were presented for chronic foreleg lameness. Pain was elicited on hyperextension of the shoulder joint. Of affected shoulder joints, 57% showed signs of DJD. Palpation of these shoulders under anesthesia revealed a craniocaudal or mediolateral drawer sign or both. Arthroscopy was most helpful in detecting signs of instability.

a. Karl STORZ Veterinary Endoscope; Tuttlingen. Germany
b. Olympus France; Scope. Rungis, France
c. OTV-S3 Olympus camera; Scope. Rungis, France
d. Sony color printer manigraph; Sony. Paris, France
e. U-Matic videocassette recorder VO-7630: Sony. Paris, France
f. Digivideo-system Karl Storz; Tuttlingen, Germany

References

1. Brinker WO. Piermatei DL. Flo G. Handbook of small animal orthopedics and fracture treatment. 1st ed. Philadelphia: WB Saunders. 1983: 369-71.

2. DeAngelis MP. Schwartz A. Surgical correction of cranial dislocation of the scapulohumeral joint in a dog. J Am Vet Med Assoc 1970;4:435-8.

3. Alexander JE. Open reduction and fixation of shoulder luxation. Sin Anim Clin 1962;7:379-83.

4. Vaughan LC. Dislocation of the shoulder joint in the dog and cat. Sm Anim Pract 1967:8:45-8.

5. DeAngelis MP. The thoracic limb: the shoulder joint. In: Bojrab MJ. ed. Current techniques in small animal surgery. 1st ed. Philadelphia: Lea & Febiger, 1975:499-504.

6. Hohn RB. Craig E. Anderson WD. Thoracic limb. In: Bojrab MJ. ed. Current techniques in small animal surgery. 2nd ed. Philadelphia: Lea & Febiger, 1983:726-33.

7. Holm RB, Rosen H. Surgical stabilization of recurrent shoulder luxation. Vet Clin Am 1971;3:537-48.

8. Wolff EF. Transposition of the biceps brachii tendon to repair luxation of the canine shoulder joint. Vet Med Sin Anim Clin 1976;69:51-3.

9. Craig E. Hohn RB. Anderson WD. Surgical stabilization of traumatic medial shoulder dislocation. J Am Anim Hosp Assoc 1980;1:93-102.

10. Lippincott CL. Reefing of the shoulder joint; a technique to surgically restore the integrity of a luxated scapulohumeral articulation in the dog. Vet Med Sm Anim Clin 1971:66:695-702.

11. Bennett D. Campbell JR. Unusual soft tissue orthopedic problems in the dog. J Anim Pract 1979:20:27-39.

12. Bedford PGC. Dislocation of the shoulder joint with fracture of the humerus in a cat. J Anim Pract 1969;10:519-22.

13. Vasseur PB. Clinical results of surgical correction of shoulder luxation in dogs. J Am Vet Med Assoc 1983:182:503-5.

14. Kavit AY, Roseann P. Surgical correction of scapulohumeral luxation in a dog. J Am Vet Med Assoc 1968;153:180-1.

15. Campbell JR. Shoulder lameness in the dog. J Sm Anim Pract 1968;9: 189-98.

16. Leighton RI_ Kagan KG. Repair of medial shoulder luxation in dogs. Mod Vet Pract 1975;57:604-6.

17. Leighton RL. Kagan KG. Surgical repair of lateral shoulder luxation. Mod Vet Pract 1976;57:102-3.

18. Parker RB, Schubert TA. Repair of ligamentous joint injuries in three dogs using spiked washers. J Am Anim Hosp Assoc 1981;17:45-50.

19. Herron MR. Osteochondral transplants for partial joint replacement. J Am Anim Hosp Assoc 1976;12:838-40.

20. Kinzel GL. Van Sickle DC, Hillsburry BM. Preliminary study of the in vivo motion of the canine shoulder. Am J Vet Res 1976;37:1505-10.

21. Vasseur PB. Moore D. Brown SA, et al. Stability of the canine shoulder joint: an in vitro analysis. Am J Vet Res 1982;43:352-5.

22. Vasseur PB, Pool RR, Klein K. et al. Effects of tendon transfer on the canine scapulohumeral joint. Am J Vet Res 1983;44:811-5.

23. Puglisi TA. Canine humeral joint instability. Comp Cont Ed 1986:8:593- 601.741-50.

24. Puglisi TA, Tangner CH. Green RW. et al. Stress radiography of the canine humeral joint. J Am Anim Hosp Assoc 1988;24:235-41.

25. Brinker WO. Piermattei DL, Flo G. Handbook of small animal orthopedics and fracture treatment. 2nd ed. Philadelphia: WB Saunders, 1990:477-81.

26. Duhautois B. L'épaule instable douloureuse : a propos de 15 cas. Prat Med Chir 1995;30:55-70.

27. Bardet JF. Subluxation de l'épaule chez un greyhound : diagnostic arthroscopique et traitement. Prat Neol Chir 1995;1329:17-22.

28. Person MW. Arthroscopy of the canine shoulder joint. Comp Cont Ed 1986;8:537-48.

29. Goring RL, Price C. Arthroscopical examination of the canine scapulohumeral joint. J Am Anim Hosp Assoc 1987;23:551-5.

30. Person MW. Arthroscopic treatment of osteochondritis dissecans in the canine shoulder. Vet Surg 1989;18:175-89.

31. Van Bree H. Van Ryssen B. Desmitt H. Osteochondritis lesions of the canine shoulder: correlation of positive contrast arthrography and arthroscopy. Vet Radiol 1992;33:342-7.

32. Van Ryssen B. Van Bree H. Vyt P. Arthroscopy of the shoulder joint in the dog. J Am Anim Hosp Assoc 1993;29:101-5.

33. Bardet JF. Arthroscopie diagnostique de l'épaule chez le chien. Prat Med Chir 1995;30:47-54.

34. Bardet JF. Traitement de l'osteochondrite dissecante de l'épaule sous arthroscopie chez le chien. Etude retrospective de 30 cas. Prat Med Chir 1996;30:685-94.

35. Goring RL, Beale BS. Failure of arthroscopic dislodgment of cartilage flaps for treatment of ostenchondritis dissecans in the shoulder joint of two racing greyhounds. J Am Anim Hosp Assoc 1987;26:423-6.

36. Neer CS. Shoulder reconstruction. Philadelphia: WB Saunders. 1990: 274-6.

37. Rockwood CA, Matsen FA. The shoulder. Philadelphia: WB Saunders. 1990:526-30.

38. Flatow EL, Pollack RC. Shoulder instability. In: Orthopedic knowledge update 5. Rosemont, IL: American Academy of Orthopedic Surgeons, 1996:233-43.

39. Payne LZ. Altcheck DW. The surgical treatment of anterior shoulder instability. Velin Sport Med 1995:14:863-83.

40. Speer KP. Anatomy and pathomechanics of shoulder instability. Chin Sport Med 1995;14:751-60.

41. Gusmer RB. Potter HG. Imaging of shoulder instability. Clin Sport Med 1995;14:777-95.

42. Johnson LL. Diagnostic and surgical arthroscopy of the shoulder. St. Louis: Mosby, 1993:425-528.

43. Snyder SJ. Shoulder arthroscopy. New York: McGraw-I till. 1994.

44. Wall MS. O'Brien Si. Arthroscopic evaluation of the unstable shoulder. Clin Sport Med 1995;14:817-39.

45. Caspari RB. Beach WR. Arthroscopy: how effective is it in making the major diagnosis? In: Matsen FA, ed. The shoulder: a balance of mobility and stability. Rosemont, IL: American Academy of Orthopedic Surgeons. 1993:369-78.

46. Evans HE, Christensen GC. Miller's anatomy of the dog. Philadelphia: WB Saunders. 1979:241-3.

47. Matsen FA, Harryman D. Sidles J. Mechanics of glenohumeral instability. Clin Sports Med 1991;10:783-8.

48. Cooper DE, Arnoczky SP. O'Brien SJ, et al. Anatomy, histology and vascularity of the glenoid labrum. J Bone Joint Surg 1992;74A:46-52.

49. Levick JR. Joint pressure-volume studies: their importance. design and interpretation. J Rheumatol 1983;21:353-7.

50. Cofield RH, Nessler JP. Wainsthal R. Diagnosis of shoulder instability by examination under anesthesia. Clin Orthop 1993;291:45-53.

51. Lepitt S. Matsen F. Mechanisms of gleno-humeral joint stability. Clin Orthop 1993;291:20-8.

52. Pollock RG, Flatow EL. Bigliani LH, et al. Shoulder biomechanics and repetitive motion. In: Gordon SL, Blair Si, eds. Repetitive motion disorders of the upper extremity. Rosemont, IL: American Academy of Orthopedic Surgeons. 1995:145-60.

53. Jerosch J. Clahsen H, Grosse-Hackmann A. et al. Effects of proprioceptive fibers in the joint capsule tissue in stabilizing the glenohumeral joint. Orthop Trans 1993;16:773.

54. Uthoff HK, Sarkar K. The effect of aging on the soft tissues of the shoulder. In: Mat sen FA, ed. The shoulder: a balance of mobility and stability. Rosemont. IL: American Academy of Orthopedic Surgeons, 1993:269-78.

55. Canoso JJ. Bursae. tendon and ligaments. Clin Rheum Dis 1981;7: 189-221.

56. Farquhar T. Cartilage bone and biomechanics. In: Proceed. Int Hip Dysplasia and Osteoarthritis Sem. Ithaca. NY: Aug 3-4.1996:22.

57. Mow VC. Setton LA, Guilak F. Ratcliffe A. Mechanical factors in articular cartilage and their role in osteoarthritis. In: Kuettner KE, Goldberg VM. eds. Osteoarthritis disorders. Rosemont. IL: American Academy of Orthopedic Surgeons. 1995:147-71.

58. Radin EL, Schaffer M. Gibson G. et al. Osteoarthritis as the result of repetitive trauma. In: Kuettner KE. Colberg VM, eds. Osteoarthritis disorders. Rosemont. IL: American Academy of Orthopedic Surgeons. 1995:197-203.

59. Flo GL, Meddleton D. Mineralization of the supraspinatous tendon in dogs. J Am Vet Med Assoc 1990;197:95-7.

60. Kriegleder H. Mineralization of the supraspinatous tendon: clinical observations in seven dogs. Vet Comp Orthop Trauma 1995:8:91-7.

61. Muir P. Goldsmith SE. Rothwell TL. et al. Calcifying tendinopathy of the biceps brachii in a dog. J Ant Vet Med Assoc 1992;201:1747.

62. Burkhead WZ. The biceps tendon. In: Rochwood CA, Matsen FA. eds. The shoulder. Philadelphia: WB Saunders, 1990:791-836.

63. Bardet JF. Arthroscopy of the shoulder joint in dogs. lit: Proceed. 8th Ann European Soc of Vet Ortho and Traumatol Meeting. Munich: April 19-21,1996:35-43.