The consultation should start with an exhaustive history including personal and sporting background. For either an isolated episode or recurrent instability, age at the time of the first episode as well as mechanism of injury should be recorded [5]. Whether or not it was reduced at the hospital or it resolved spontaneously, as well as any neurological complication, are also important details that can give an idea of the severity of the first episode. In chronic instability cases, one should include the number of dislocations, activities during which it happens and type of reduction needed.

Physical examination should always be performed bilaterally to provide comparison. It should include visual inspection, palpation, active and passive range of motion and motor and sensory testing. Then, a general shoulder examination can be performed, including specific tests selected according to clinical suspicion. The diagnosis of occult instability should be always considered in young athletes with a painful shoulder [8].

The aim of this review is to analyze the characteristics of physical examination in different laxity and instability patterns.

The aim of this test is to demonstrate the amount of antero-posterior translation of the humeral head on the glenoid. Silliman and Hawkins first described it in 1991 [18] and it can be performed with the patient sitting or supine [19]. With the patient sitting, the examiner stabilizes the scapula by placing one hand over the top of the shoulder while the other is placed on the proximal arm. It is important to load the humeral head into the glenoid before starting anterior or posterior displacement. This test has not been validated using biomechanical techniques. Tzannes and Murrell found a sensitivity of 50% for anterior instability and 14% for posterior instability while the specificity was of 100% for both of them [19].

Gerber and Ganz described anterior and posterior drawer tests in 1984 [20]. Examination is performed with the patient supine and the examiner to the side.

The posterior drawer test is performed by holding the patient´s wrist [20] or forearm [1] with one hand and placing the other hand over the patient´s shoulder so that the thumb is in the front and the fingers in the back. The thumb should be placed over the humeral head while applying a posteriorly directed force. The arm should be forwardly flexed at the same time in order to allow the head to subluxate posteriorly. The displacement is measured as explained before. Saha et al. [21] described the “zero unpacked position” (arm elevated 45º to 60º) as the position in which shoulder has the most mobility. These authors recommend performing the posterior drawer test in this position to obtain more reliable results.

For the anterior drawer test the examiner has to control scapular rotation. The arm is abducted 80º to 120º, flexed 0º to 20º and held in 0º to 30º of external rotation [20]. In this case, one hand stabilizes the scapula with the thumb in the coracoid and the fingers in the scapular spine. The other hand creates an anteriorly directed force to provoke and anterior translation of the humeral head. McFarland et al. [1] described another technique with the arm in 40º of abduction and slight internal rotation. One hand is placed on the arm and the other in the wrist, while applying an axial load to the shoulder. In this position, the authors recommend the force to be applied in the whole arm and not only in the humeral head (Fig. 1).

Fig. (1). Drawer test. One hand stabilizes the scapula with the thumb in the coracoid and the fingers in the scapular spine. The other hand creates an anteriorly directed force to provoke and anterior translation of the humeral head.

It has been described that grade II displacement could be considered normal among young athletic populations and that asymmetry of the shoulders should not be used as a criterion for diagnosis of shoulder laxity [22].

Neer and Foster [23] described the sulcus test in 1980 in the context of a multidirectional instability study and it has remained as a way to examine inferior laxity. Later biomechanical investigation has demonstrated that it evaluates the superior glenohumeral ligament [24, 25]. This examination can be performed with the patient standing, sitting or supine. McFarland et al. have suggested that more reliable data can be obtained with the patient sitting and the arm by the side [1]. The elbow is grasped and pulled inferiorly (Fig. 2). If the test is positive, a sulcus appears in the subacromial region as the humeral head translates in the inferior direction [19]. It has to be repeated twice, first with the arm in neutral rotation and secondly with the arm in external rotation. Inferior translation should be the same in both positions. In the case that we find an increased translation with the arm in external rotation a lesion of the rotator interval should be considered [25, 26] (Fig. 2).

Fig. (2). Sulcus sign. The arm is held first in neutral rotation (a) and secondly in external rotation (b). The elbow is grasped and pulled inferiorly. If the test is positive, a sulcus appears in the subacromial region as the humeral head translates in the inferior direction.

The amount of displacement can be measured in centimeters. Typically, grade I would imply a displacement of less than 1.5 cm, grade II between 1.5 and 2 cm and grade III more than 2 cm. McFarland [1] has suggested grouping these values as low (grade I) and high (grades II and III). Tzannes and Murrell [27] found that a positive sulcus sign of more than 2 cm had a sensitivity of 28% and a specificity of 97%. However, other authors have demonstrated a wide range of inferior laxity in asymptomatic patients, suggesting that there is no absolute degree of inferior translation that defines inferior instability [1].

This test was reported by Dr. O.J Gagey and Dr. N. Gagey in 2001 to assess laxity in the inferior glenohumeral complex [28]. It is performed with the examiner standing behind the patient using one arm to stabilize the scapula (Fig. 3). The patient´s arm is abducted until the scapula is felt to start moving. The test is positive if the range of passive abduction is over 105º. The authors reported a 15% of patients in which passive abduction was limited by apprehension what could somehow contribute to the diagnosis of instability [28]. At present, this test has evolved to the “comparative hyperabduction test” in which both shoulders are examined and three items should be noted in order to consider the result positive. First, it has to reproduce patient´s deep pain; Secondly, it has to be asymmetrical when compared to the contralateral side (>20º) and third; a soft end point should not be felt [8] (Fig. 3).

Fig. (3). The Hyperabduction test by Gagey is performed as shown. Passive abduction is performed until the scapula starts moving. It should be performed bilaterally to obtain a comparison.

Various tests have been described in order to examine posterior shoulder laxity. The push-pull test [29] and the Protzman test [30] could be considered modifications of the posterior load and shift test. Clarnette and Miniaci [31] also described a posterior subluxation test performed in internal rotation, adduction and flexion that has not yet been validated.

Anterior glenohumeral instability is the most frequent problem encountered [5]. Thus, most of the tests have been described in order to diagnose it. Same as in laxity examination, the testing methods and what is considered a positive test are still controversial [32].

3.1. The Apprehension Sign/ Augmentation test

Since Rowe and Zarins [33] first described this test for anterior shoulder instability many variations have been suggested. It can be performed either standing or supine. In the standing position, the examiner places the arm in external rotation and abduction of 90º with one hand while the other applies a gentle pressure anteriorly. Farber et al. [3] reported more reliable results when including this test in the general examination routine, holding both arms by the forearm and exploring both at the same time without exerting any pressure on the shoulder. Another way to perform the apprehension test is with the patient in supine position, placed on the side of the table keeping the scapula supported (Fig. 4). The shoulder is held in a position of 90º of abduction with the elbow at 90º of flexion and the examiner´s knee supporting the elbow to prevent extension of the shoulder [32]. From there, external rotation is applied until the patient feels pain or becomes apprehensive. Apprehension used as a criterion has demonstrated to achieve better inter-observer agreement whereas pain might overestimate the number of positive results [27, 32, 34]. When pain is felt, the examiner should ask the patient to localize it in the posterior, anterior or lateral aspect of the shoulder. The specificity of this test has been reported as 98.9% with low sensitivity of 52.8% [32].

Several variations of this test have been described. In the “augmentation test” [35], the “crank test” [35] and the “fulcrum test” [29] an anterior force is applied in the position of maximum apprehension.

Fig. (4). Anterior apprehension test and relocation test. The shoulder is held in a position of 90º of abduction with the elbow at 90º of flexion and external rotation is applied from there until the patient feels pain or becomes apprehensive (a & b). Then, a posteriorly directed force is applied on the anterior aspect of the shoulder (c). The examiner should notice if the patient experiences relief of apprehension or just relief of pain in order to consider it positive.

Posterior instability is much less common than anterior instability, comprising between 2% and 10% of all reported instability cases [39]. This means it can be missed in the absence of a clear traumatic dislocation episode such as in sporting activities, a seizure episode or an electrical shock that gives rise to suspicion. The mechanism of injury or situations that reproduce the symptoms should be recorded and might orientate the clinician to this type of instability, as they usually coexist with other shoulder problems such as impingement or labral tears [39]. In young athletes, as mentioned before, laxity should not be mistaken for instability. Although examination at first glance could seem benign, Pollock et al. [40] described that patients are commonly tender to palpation along the posterior joint line, possibly from synovitis provoked by multiple episodes of instability. Von Raebox et al. have also reported the presence of a small skin dimple or tether over the posteromedial deltoid of both shoulders in patients with posterior positional dislocation [41].

Many provocative maneuvers have been described. The posterior apprehension sign was described as a test in which the arm was pushed posteriorly while being held in a position of flexion, adduction and internal rotation [29]. However, not all the patients with confirmed posterior instability feel the symptoms in the same position [21]. Later, O´Driscoll and Evans described the “posterior apprehension test for pain” that consisted in reproducing a Kennedy-Hawkins impingement sign. In the cases in which pain was elicited, a subacromial injection was performed to rule out glenohumeral symptoms [42].

Fig. (5). Jerk test. With the patient seated, the shoulder is stabilized with one hand. With the other hand, the patient grasps the elbow and abducts the arm to 90º while internally rotating and applying an axial load directed to the shoulder (a & b). Keeping the axial load, the arm is adducted until a sudden clunk or jerk appears (c & d).

The Jerk test was described by Matsen et al. [29] The arm is brought into flexion, adduction and internal rotation with the elbow flexed as well. Then, the examiner applies an axial force along the humerus to induce a posterior subluxation (Fig. 5). The Kim test is performed with the patient sitting (Fig. 6). The examiner holds the elbow with one hand and keeps the other over the patient’s biceps. Then, the arm is then passively flexed 45º while applying a downward and posterior force to the upper arm and an axial load to the elbow [43]. When performed together, they have been reported to provide a sensitivity of 97% for detecting postero-inferior labral tears [43].

Fig. (6). Kim test. (a) While the patient seated, the examiner grabs the elbow with one hand and the biceps area with the other (b). The arm is then passively elevated while applying a posterior force to the upper arm and an axial load to the elbow. A positive result would mean pain or subluxation.