Physical Examination of the Shoulder




key words

shoulder examination, stability testing, strength testing, test sensitivity, test specificity

 




Introduction


The shoulder girdle allows for a large degree of motion in multiple planes, with the glenohumeral joint being the most mobile joint in the body. The tradeoff for this freedom of motion is a relative lack of stability, which makes the shoulder girdle susceptible to an array of injuries. A number of physical examination maneuvers have been developed to assist examiners in diagnosing shoulder problems. Performing these maneuvers accurately and understanding their reliability and validity are paramount to a proper shoulder examination. In this chapter, we review common shoulder examination maneuvers, identifying the original descriptions and presenting research examining the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the various tests.




Examination of the Shoulder


A thorough examination of shoulder symptoms should include the cervical spine, contralateral shoulder, elbow, trunk, and upper-limb neurovascular structures. We limit our focus to the shoulder girdle, which includes the sternoclavicular, acromioclavicular (AC), glenohumeral, and scapulothoracic (ST) joints. Doing the basic aspects of a musculoskeletal examination is especially important in the shoulder: The key to performing a good shoulder examination is to develop a system in which the patient is prepped so you can (1) see the shoulders; (2) compare both sides; (3) do a neurovascular examination; and (4) consider the joint above, which in this case is the cervical spine.


Inspection


The first step of shoulder examination is to have the patient undress so that both shoulders can be examined and compared. For men, this is accomplished by taking off the shirt, and for women a sports bra or a gown worn around the thorax can suffice ( Fig. 4.1 ). The patient should be examined from the front and the back, where elements such as muscle bulk and scapular positioning can be easily observed. Posture should be observed in both the seated and standing positions and from different angles. Scars, atrophy, swelling, ecchymosis, erythema, rashes, deformities, shoulder heights, and scapular positioning should be evaluated. Posture in the standing and seated positions should be observed for a forward set, protracted head, and rounded shoulders (humeral internal rotation and scapular protraction), which will cause functional narrowing of the subacromial space. Scapular winging may be seen and can be accentuated by muscle activation ( Fig. 4.2 ). Observing the shoulder girdle from the back of the patient during arm flexion and abduction may reveal altered movement of the scapula secondary to muscle weakness or imbalances in flexibilities.




Figure 4.1


It is helpful to dress the patient so that both shoulders can be seen completely, allowing side-to-side comparison.



Figure 4.2


Winging of the scapula.

(Reproduced with permission from Hawkins RJ, Bokor DJ: Clinical evaluation of shoulder problems. In Rockwood CA, Matsen FA (eds), The Shoulder , 2nd edn. Philadelphia: WB Saunders, 1998, p. 172.)


Palpation


The superficial structures that should be evaluated are the sternal notch, sternoclavicular joint, clavicle, AC joint, long head of the biceps tendon, subacromial bursae, greater and lesser tuberosities of the humerus, coracoid process, supraclavicular fossa, and spine of the scapula with its borders ( Fig. 4.3 ). The AC joint is superficial and is identified with palpation of the clavicle and spine of the scapula until they meet laterally. The long head of the biceps is anterior, between the lesser and greater humeral tuberosities, and is difficult to palpate because of the large deltoid muscle. By externally rotating the arm and flexing and extending the elbow, the examiner may be able to feel the tendon moving in the anterior shoulder. The cervical spine and trapezius should be palpated if the patient has neck pain.




Figure 4.3


Surface anatomy of shoulder region. AC, Acromioclavicular; SC, sternoclavicular.


The attachments of the muscles to the scapula are noted in Figure 4.4 . When indicated, the axilla should be evaluated for masses, lymph nodes, and palpation of the muscles. The pectoralis major lies anterior and covers the pectoralis minor, which is difficult to palpate. The pectoralis minor muscle, when tight, has been implicated in an internally rotated and protracted scapula. The latissimus dorsi forms the posterior border and may occasionally be torn, especially in baseball pitchers.




Figure 4.4


Muscular anatomy about the scapula. A, Anterior view. B, Posterior view.

(Adapted from Jobe CM. Gross anatomy of the shoulder. In: Rockwood CA, Matsen FA, eds. The Shoulder. 2nd ed. Philadelphia: W.B. Saunders; 1998:44.)


Range of Motion Testing


Active range of motion testing is usually performed first to allow the patient to feel comfortable and avoid painful positions. Passive motion testing can then be performed to isolate motions for accurate evaluation. The active and passive range of motion of both sides should be compared. The planes of shoulder girdle motion include forward flexion, extension, internal/external rotation, abduction/adduction, and a combination called circumduction . Range of motion is noted by degrees from a reference position; usually the anatomic position is used without scapular fixation unless otherwise specified.


The first measure of shoulder motion should be elevation of the arm. Elevation can be performed with the arm in abduction or flexion. Abduction of the arm can be performed in the plane of the body but is best performed in the “scapular plane,” which is approximately 30 degrees in front of the plane of the body ( Fig. 4.5 ). Both forward flexion and abduction are typically at least 160 degrees but may exceed this in flexible athletes. The next motions to evaluate are shoulder rotations. The neutral position is with the arm and forearm in the horizontal plane ( Fig. 4.6A ). The first is done with the arm abducted 90 degrees and typically supported by the examiner holding the elbow. External rotation ( Fig. 4.6B ) and internal rotation ( Fig. 4.6C ) at this elevation typically include not only motion of the ST articulation but also the glenohumeral joint. Next, external rotation with the arm at the side should be compared with that of the opposite extremity.




Figure 4.5


Plane of the scapula is approximately 30 degrees in front of the plane of the body.

(Redrawn from McFarland EG: TK Kim, HB Park, G El Rassi, H Gill, E Keyurapan: Examination of the Shoulder: The Complete Guide, New York, Thieme, 2006, pp 162-212 Fig 2.4.)



Figure 4.6


When examining the shoulders for rotation, the starting position is shown ( A ) with the arms in a neutral position. B, External rotation and ( C ) internal rotation can be measured from the side.






Internal rotation cannot be accurately measured with the arm at the side in this position because the trunk impedes the motion. Internal rotation of the shoulder can be performed by asking the patient to place the arms up the back with the thumbs up ( Fig. 4.7 ). The landmarks typically used for this measure are the hip, buttock, sacrum, L1 body, lower border of the scapula around T8, and prominent C4 vertebral spinous process. This method of measurement can be reproducible for one individual, but the relationship of the thumb tip to various vertebral levels has not been shown to be accurate or reproducible. One functional measure of internal rotation is the Apley scratch test, but it is not practical because most people cannot perform the maneuver ( Fig. 4.8 ).




Figure 4.7


Internal rotation of the arm up the back is performed as pictured here.



Figure 4.8


The Apley scratch test is a measure of several joint ranges of motion and not just the shoulder.


Normal values of active range of motion for the shoulder joint are shown in Table 4.1 . When evaluating shoulder motion, it is sometimes important to measure glenohumeral motion while preventing ST motion. Isolating glenohumeral motion with the arm abducted 90 degrees involves externally or internally rotating the arm until scapular motion is perceived manually and visually. Internal and external rotation from this position can vary greatly, particularly in overhead athletes. Generally, glenohumeral external rotation is 90 degrees or more, and internal rotation is 0 to 30 degrees with the arm abducted 90 degrees. External rotation with the arm at the side can be measured either as glenohumeral motion alone or combined with ST motion.



Table 4.1

Normal Active Shoulder Range of Motion

Reproduced with permission from Moore KL. The upper limb. In: Clinically Oriented Anatomy. 2nd ed. Baltimore: Williams & Wilkins; 1985.






















Position Degrees
Forward flexion/elevation a 0–180
Extension a 0–60
Abduction a 0–180
Glenohumeral internal rotation b 0–70
Glenohumeral external rotation b 0–90

a Zero begins at the anatomic position.


b Zero begins with the humerus abducted to 90 degrees.



Muscles, Innervations, and Biomechanics


The muscles of the shoulder consist of the stabilizing rotator cuff (supraspinatus, infraspinatus, teres minor, and subscapularis; Fig. 4.9 ), trapezius, serratus anterior, rhomboids, and the prime movers (pectoralis major/minor, latissimus dorsi, teres major, triceps, biceps, and deltoid; Fig. 4.10 ). Most of the shoulder girdle is supplied by the fifth and sixth cervical roots through the upper trunk of the brachial plexus.




Figure 4.9


The rotator cuff muscles function to compress the humeral head into the glenoid and to rotate the arm.

(Reproduced with permission from Perry J. Anatomy and biomechanics of the shoulder in throwing, swimming, gymnastics, and tennis. Clin Sports Med . 1983;2(2):252.)



Figure 4.10


Prime movers about the shoulder girdle shown on magnetic resonance imaging ( A ) and illustrated ( B ): 1, pectoralis major; 2, pectoralis minor; 3, first rib; 4, serratus anterior; 5, second rib; 6, third rib; 7, rhomboid; 8, trapezius; 9, subscapularis; 10, infraspinatus; 11, deltoid.

(Adapted from Jobe CM. Gross anatomy of the shoulder. In: Rockwood CA, Matsen FA, eds. The Shoulder. 2nd ed. Philadelphia: W.B. Saunders; 1998:43.)


The suprascapular nerve (C5–C6) innervates the supraspinatus and infraspinatus, which originate from the supraspinatus and infraspinatus fossa, respectively. The supraspinatus inserts onto the superior facet of the greater tuberosity, whereas the infraspinatus inserts on the middle facet. The axillary nerve (C5–C6) innervates the deltoid and teres minor. The deltoid originates from the lateral third of the clavicle and scapular spine and includes the AC joint; it inserts onto the deltoid tuberosity of the humerus. The teres minor originates from the superior lateral portion of the scapula and inserts onto the inferior aspect of the greater tuberosity. The subscapularis is innervated by the nerve to the subscapularis (upper and lower), composed of the cervical 5, 6, and 7 roots. It originates from the anterior portion of the scapula (subscapularis fossa) and inserts onto the lesser tuberosity of the humerus.


The trapezius contains three portions—upper, middle, and lower. It is innervated by the spinal accessory, 11th cranial nerve (C3–C4). It has a vast origin from the occipital protuberance and superior nuchal line superiorly to the 12th thoracic vertebra inferiorly. It inserts onto the lateral third of the clavicle, acromion, and spine of the scapula. The long thoracic nerve (C5–C7) innervates the serratus anterior. It originates from the lateral portions of the first eight ribs and inserts onto the anterior surface of the medial border of the scapula. The rhomboids include the major and minor divisions and are innervated by the dorsal scapular nerve (C5). They originate from the ligamentum nuchae and spinous processes from C7 to T5 and insert onto the medial border of the scapula from the scapular spine to the inferior angle.


The pectoralis major has two components, the clavicular and sternocostal divisions, which are innervated by the lateral and medial pectoral nerves (clavicular, C5–C6 and sternocostal, C7–T1). The pectoralis minor is also innervated by these nerves (C6–C8). The major originates from the medial portion of the clavicle, sternum, and second to sixth ribs and inserts onto the humeral lateral lip of the intertubercular groove. The minor originates from ribs 3 to 5 and inserts onto the medial coracoid. The latissimus dorsi is supplied by the thoracodorsal nerve (C6–C8) and has a large origin of the spinous processes of T6 to the sacrum, the thoracolumbar fascia, iliac crest, and the caudal three ribs while inserting onto the floor of the intertubercular groove. The teres major is supplied by the lower subscapular nerve (C6–C7). It originates on the dorsal surface of the inferior angle of the scapula and inserts onto the medial lip of the intertubercular groove. The triceps has three heads, the long, lateral, and medial, which are supplied by the radial nerve (C6–C8). The long head originates from the infraglenoid tubercle of the scapula, and the lateral and medial heads originate from the posterior surface of the humerus superior and inferior to the spiral groove, respectively. They insert onto the proximal ulna (olecranon). The biceps comprises the long and short heads innervated by the musculocutaneous nerve (C5–C6). The long head originates from the supraglenoid tubercle of the scapula and the short head from the coracoid process of the scapula, and both insert onto the radial tuberosity and flow into the bicipital aponeurosis.


Scapular Biomechanics


Saha has discussed three layers of muscles that stabilize the scapula and assist in force production from the musculature. The rotator cuff muscles (supraspinatus, infraspinatus, subscapularis, and teres minor) are the inner layer; these muscles serve first to provide compressive force of the humeral head into the glenoid and secondly to provide rotation of the arm. The middle layer comprises the teres major, pectoralis major, the latissimus dorsi, and the short fibers of the anterior and posterior deltoid. The superficial layer is the triceps, long head of the biceps, coracobrachialis, and superficial fibers of the anterior and posterior deltoid. The trapezius, rhomboids, and serratus anterior provide stabilizing forces because the scapula lacks rigid, bony fixation.


The upper trapezius, levator scapula, and superior serratus anterior elevate the scapula; the pectoralis minor and major and latissimus dorsi depress the scapula; the serratus anterior, pectoralis minor, and levator scapula protract the scapula; the trapezius, rhomboids, and latissimus dorsi retract the scapula; the superior and inferior portions of the trapezius and inferior portion of the serratus anterior cause lateral scapular rotation; and the levator scapula, rhomboids, pectoralis minor, and major and latissimus dorsi cause medial scapular rotation. These muscles fire in a coordinated fashion to perform the resultant actions in a smooth and effective manner, known as force couples .


Proper positioning of the scapula throughout motion allows the muscles associated with the scapula to have the appropriate length–tension relationships for the greatest efficiency of limb positioning. With the scapula stabilized, the glenoid can be maintained for humeral motion upon it. As the humerus is abducted, the glenohumeral to ST range of motion occurs at approximately a 2 : 1 ratio. This ratio changes through the arc of motion; that is, the 2 : 1 ratio is not constant throughout the entire range of motion. In the initial portion of abduction, glenohumeral motion predominates, and the ratio has been found to be 4.4 degrees of glenohumeral motion for every degree of ST motion. As the shoulder moves above 90 degrees of abduction, this ratio becomes 1.1 degrees of glenohumeral to 1 degree of ST motion. This scapular rotation during abduction also elevates the acromion, which has been postulated to help prevent impingement of the rotator cuff upon the acromion.


Although the muscles are the dynamic stabilizers, the static stabilizers of the ligaments and joint capsule should not be forgotten ( Fig. 4.11 ). The primary stabilizer of anterior translation with the arm abducted to 90 degrees is the anterior band of the inferior glenohumeral ligament complex (IGHLC). With the arm in lesser degrees of abduction, the middle glenohumeral ligament restricts external rotation. Limitation of posterior translation is by the posterior band of the IGHLC, whereas inferior translation is limited by the inferior capsule and, at the top of the shoulder, the superior glenohumeral ligament ( Fig. 4.12 ). Recently, it has been noted that the inferior glenohumeral ligament also contributes to limitation of inferior motion with the arm abducted.




Figure 4.11


Anatomy of the glenohumeral ligaments. AB, anterior band; B, long head of biceps; IGHLC, inferior glenohumeral ligament complex; MGHL, middle glenohumeral ligament; PB, posterior band; PC, posterior capsule; SGHL, superior glenohumeral ligament.

(From Bowen, MK, Warren RF: Ligamentous control of shoulder stability based on selective cutting and static translation experiments. Clin Sports Med. 1991;10:763.)



Figure 4.12


The superior glenohumeral ligament (SGHL) is the primary restraint to inferior translation. AB, Anterior band; MGHL, middle glenohumeral ligament; PB, posterior band.

(Reproduced with permission from Bowen MK, Warren, RF. Ligamentous control of shoulder stability based on selective cutting and static translation experiments. Clin Sport Med. 1991;10:769.)


Tests of Rotator Cuff Strength and Integrity


See Table 4.2 .



Table 4.2

Tests of Rotator Cuff Strength and Integrity
























Test Description Clinical Utility
Empty can test The supraspinatus test is first performed by assessing the deltoid with the arm at 90 degrees of abduction and neutral rotation. The shoulder is then internally rotated and angled forward 30 degrees: the thumb should be pointing toward the floor. Muscle testing against resistance is then performed.


  • Naredo et al.




    • For detecting supraspinatus lesion:




      • Sensitivity: 79.3%



      • Specificity: 50%




    • For detecting supraspinatus tendonitis:




      • Sensitivity: 77.2%



      • Specificity: 38.4%




    • For detecting supraspinatus tears:




      • Sensitivity: 18.7%



      • Specificity: 100%





  • Park et al.




    • For detecting partial-thickness RC tear:




      • Sensitivity: 32.1%



      • Specificity: 67.8%




    • For detecting full-thickness RC tear:




      • Sensitivity: 52.6%



      • Specificity: 82.4%



Full can test The examiner abducts the arm at 90 degrees of abduction and neutral rotation. The shoulder is then externally rotated with thumb pointing toward the roof. Muscle testing against the resistance is then performed.


  • Itoi et al.




    • For detecting partial-thickness RC tear:




      • Sensitivity: 83%



      • Specificity: 53%



      • Accuracy: 78%



Drop arm test The examiner abducts the patient’s shoulder to 90 degrees and then asks the patient to slowly lower the arm to the side in the same arc of movement. A positive test result is indicated if the patient is unable to return the arm to the side slowly or has severe pain when attempting to do so.


  • Bryant et al.




    • For detecting RC tear:




      • Positive predictive value: 100%



      • Sensitivity: 10%



Patte test The examiner supports the patient’s elbow in 90 degrees of forward elevation in the plane of the scapula while the patient is asked to rotate the arm laterally to compare the strength of lateral rotation. Jobe and Patte maneuvers can produce three types of responses: (1) absence of pain, indicating that the tested tendon is normal; (2) the ability to resist despite pain, denoting tendonitis; or (3) the inability to resist with gradual lowering of the arm or forearm, indicating tendon rupture.


  • Naredo et al.




    • For detecting infraspinatus lesions:




      • Sensitivity: 70.5%



      • Specificity: 90%




    • For detecting infraspinatus tendonitis:




      • Sensitivity: 57.1%



      • Specificity: 70.8%




    • For detecting infraspinatus tears:




      • Sensitivity: 36.3%



      • Specificity: 95%




RC, rotator cuff.


Dynamic stability of the glenohumeral joint is provided by contraction of the rotator cuff and, to a lesser degree, the long head of the biceps. These tendons increase compression across the glenohumeral joint and dynamically maintain the position of the humeral head within the glenoid. As the load on the arm increases, these muscles increase the contraction necessary to keep the humeral head in the socket.


Jenp and coworkers used electromyography to detect the most specific positions for activating particular rotator cuff muscles. The supraspinatus could not be effectively isolated from the deltoid muscle when resisting abduction of the arm, but it is typically tested with the arm elevated 90 degrees with the thumb in internal, neutral, or external rotation. With the arm in this position and the thumb in internal rotation, this test is known as the “Jobe test.” However, subsequent study has found that the test has equal validity whether the thumb is pointing down, neutral, or up. The subscapularis’ greatest activation was with the arm in the scapular plane at 90 degrees of elevation and neutral humeral rotation. The infraspinatus is best tested with the arms at the side ( Fig. 4.13 ). The teres minor is best tested with the arm abducted 90 degrees and externally rotated 90 degrees ( Fig. 4.14 ).




Figure 4.13


The infraspinatus is best assessed by testing external rotation with the arms at the side. Arrows show direction of examiner’s force.



Figure 4.14


Patte test for testing teres minor and infraspinatus. Arrow shows direction of examiner’s force.


Author comment: You can have a complete tear of the rotator cuff but have complete range of motion. The difference between a shoulder with an intact rotator cuff and a torn rotator cuff is that the latter will be weak with abduction and external rotation.


Empty Can Test


Jobe described the empty can test—also known as the supraspinatus test—to help in evaluating the strength of the supraspinatus muscle ( Fig. 4.15 ). Jobe originally described the test as follows:




The supraspinatus test is first performed by assessing the deltoid with the arm at 90 degrees of abduction and neutral rotation. The shoulder is then internally rotated and angled forward 30 degrees: the thumb should be pointing toward the floor. Muscle testing against resistance is then performed ( ).




Figure 4.15


The Jobe (empty can) test is a test of the supraspinatus and deltoid muscles. The arms are abducted 90 degrees in the scapular plane with the elbows extended and the thumbs pointing down. The examiner pushes down, and a positive test result is pain or weakness. Arrows show the direction of the examiner’s force.


The test result was positive when the patient reported pain or demonstrated weakness with the arm in this position. Unfortunately, the empty can test can be painful for many patients with shoulder conditions. We recommend performing this test first with the elbows bent to avoid injuring or aggravating the shoulder. Electromyographic study has shown that, in this position, the downward force is resisted by the deltoid and the supraspinatus muscles, so this test does not isolate the supraspinatus. Malanga and associates examined the rotator cuff muscles via electromyography using two testing positions on the basis of recommendations by Jobe and Moynes and Blackburn and coworkers. They noted the supraspinatus was sufficiently activated in both positions ( Figs. 4.15 and 4.16 ).




Figure 4.16


Electromyographic studies have shown that the Jobe test can test the supraspinatus and deltoid equally to the empty can test. This position is the “full can test” and is often less painful for patients than the empty can test. Arrows show direction of examiner’s force.


The sensitivity and specificity of the Jobe test depend on the methods used for each study but also vary according to the type of rotator cuff lesion. The literature suggests that a positive Jobe test is sensitive and moderately specific for a tear of the supraspinatus tendon.


Full Can Test


The Jobe test for strength testing of the supraspinatus can be performed in the thumb-up position (see Fig. 4.16 ).


The examiner abducts the arm at 90 degrees of abduction and neutral rotation. The shoulder is then externally rotated with the thumb pointing toward the roof. Muscle testing against the resistance is then performed. A positive test result is indicated by pain, weakness, or both.


Itoi and others reported a sensitivity of 83%, specificity of 53%, and accuracy of 78% for the full can test in detecting partial-thickness rotator cuff tears.


Drop Arm Test


The drop arm test has been used to assess for rotator cuff tears, particularly of the supraspinatus. Although the original description of the drop arm test remains obscure, it has been ascribed to Codman and described by Magee as follows:




The examiner abducts the patient’s shoulder to 90 degrees and then asks the patient to slowly lower the arm to the side in the same arc of movement. A positive test is indicated if the patient is unable to return the arm to the side slowly or has severe pain when attempting to do so ( ).



Bryant and coworkers studied 53 patients with a suspicion for rotator cuff tear and compared physical examination tests to the results of MRI and ultrasonography of the shoulder. They found the drop arm test to have a 100% PPV (ie, if present, the patient has a tear) and 10% sensitivity (ie, if negative, the patient could still have a tear). It is important to realize that a positive drop arm test result can be caused by weakness of any cause, including cervical disc disease, brachial plexopathy, brachial neuritis, stroke, amyotrophic sclerosis, and many other neurologic factors.


Test of Infraspinatus and Teres Minor Integrity


See Table 4.2 .


As noted, previous electromyographic data have failed to differentiate the function of the infraspinatus and teres minor. However, the strength of the infraspinatus can best be tested with resisted external rotation with the arm at the side (see Fig. 4.13 ).


Patte Test ( )


Naredo and coworkers reported a test described by Patte in 1995 for assessing tears of the infraspinatus and teres minor (see Fig. 4.14 ). They write:




… the examiner supports the patient’s elbow in 90 degrees of forward elevation in the plane of the scapula while the patient is asked to rotate the arm laterally in order to compare the strength of lateral rotation. Jobe’s and Patte’s manoeuvres can produce three types of response: (a) absence of pain, indicating that the tested tendon is normal; (b) the ability to resist despite pain, denoting tendinitis; or (c) the inability to resist with gradual lowering of the arm or forearm, indicating tendon rupture.



Naredo and associates compared the Patte test with findings on ultrasonography and showed the test to have a sensitivity of 70.5%, specificity of 90%, PPV of 85.7%, and NPV of 70.5% for detecting infraspinatus lesions; a sensitivity of 57.1%, specificity of 70.8%, PPV of 36.3%, and NPV of 85% for detecting infraspinatus tendonitis; and a sensitivity of 36.3%, specificity of 95%, PPV of 80%, and NPV of 73% for detecting infraspinatus tears.


Tests of Subscapularis Strength


See Table 4.3 .



Table 4.3

Tests of Subscapularis Strength




















Test Description Clinical Utility
Lift-off test The hand of the affected arm is placed on the back at the mid-lumbar region, and the patient is asked to rotate the arm internally and lift the hand posteriorly off the back. A positive test result is when the patient cannot lift the hand off the back.


  • Bartsch et al. :




    • Sensitivity: 40%



    • Specificity: 79%



    • PPV: 50%



    • NPV: 71%



    • Accuracy: 66%




  • Salaffi et al. :




    • Sensitivity: 35%



    • Specificity: 75%



    • PPV: 85%



    • NPV: 21%



    • Accuracy: 64%


Lift-off lag sign In sitting position, the hand on the side of the painful shoulder is placed at the lumbar region (hand behind back). The hand is passively lifted from the lumbar spine until almost full internal rotation is reached, and the patient is asked to maintain the position actively. The test result is positive if the patient cannot maintain the position.


  • Miller et al. :




    • Sensitivity: 100%



    • Specificity: 84%



    • PPV: 28%



    • NPV: 100%




  • Bartsch et al. :




    • Sensitivity: 71%



    • Specificity: 60%



    • PPV: 45%



    • NPV: 81%



    • Accuracy: 63%


Bear hug test The bear hug test was described by Barth et al. and is performed by asking the patient to place the hand on the side of the shoulder to be tested on the opposite shoulder. The examiner then asks the patient to try to keep the hand on the shoulder while the examiner attempts to pull it off the opposite shoulder. A test result is considered positive when the patient cannot keep the hand on the shoulder and it pulls away.


  • Barth et al. :




    • Sensitivity: 60%



    • Specificity: 92%



    • PPV: 75%



    • NPV: 85%



    • Accuracy: 82%



NPV, Negative predictive value; PPV, positive predictive value.


Lift-Off Test


Muscle strength of the subscapularis can be tested with the lift-off maneuver. The test was first described by Gerber and Krushell in 1991 and was originally performed with the hand up the back ( Fig. 4.17 ). The patient was asked to lift the hand off the buttocks, and if this was not possible, then a subscapularis tendon tear was considered present. Electromyographic study has demonstrated the validity of this test for specificity of the subscapularis ( ).




Figure 4.17


The lift-off test is performed by having the patient lift the hand off the lower back as shown (arrow). If the patient cannot do this, then the test result is positive for a subscapularis tendon tear.


Lift-Off Lag Sign


A variation of the lift-off test is the lift-off lag sign. In this test, the examiner holds the elbow of the patient and lifts the hand off the midsacrum level ( Fig. 4.18A ). The patient is asked not to let the arm or forearm fall to the buttocks; a test result is considered positive if the arm falls to the buttocks or toward the floor ( Fig. 4.18B ).




Figure 4.18


A, The lift-off lag sign is performed by holding the patient’s hand away from the lower back while stabilizing the elbow. B, A positive test result is when the hand falls back to the torso and cannot stay in the starting position.




Bear Hug Test ( )


The bear hug test was described by Barth and associates and is performed by asking the patient to place the hand on the side of the shoulder to be tested on the opposite shoulder ( Fig. 4.19 ). The examiner then asks the patient to try to keep the hand on the shoulder while the examiner attempts to pull it off the opposite shoulder. A test result is considered positive when the patient cannot keep the hand on the shoulder and it pulls away.




Figure 4.19


The bear hug test is performed by having the patient place the hand of the affected shoulder on the opposite shoulder. The examiner then tries to pull the hand off the shoulder. The arrow shows direction of examiner’s force. A positive test result is when the wrist flexes or the hand can be pulled away from the shoulder.


Tests of Scapular Pathology


See Table 4.4 .



Table 4.4

Tests of Scapular Pathology
























Test Description Reliability/Validity
Lateral scapular slide test The first position of the test is with the arm relaxed at the side. The second is with the hands on the hips with the fingers anterior and the thumb posterior with approximately 10 degrees of shoulder extension. The third position is with the arms at or below 90 degrees of arm elevation with maximal internal rotation at the glenohumeral joint. These positions offer a graded challenge to the functioning of the shoulder muscles to stabilize the scapula. The final position presents a challenge to the muscles in the position of most common function at 90 degrees of shoulder elevation.


  • Odom et al.




    • With 1.5-cm difference as positive:




      • In first position:




        • Sensitivity: 28%



        • Specificity: 53%




      • In second position:




        • Sensitivity: 50%



        • Specificity: 58%




      • In third position:




        • Sensitivity: 34%



        • Specificity: 52%





    • With 1-cm difference as positive:




      • In first position:




        • Sensitivity: 35%



        • Specificity: 48%




      • In second position:




        • Sensitivity: 41%



        • Specificity: 54%




      • In third position:




        • Sensitivity: 43%



        • Specificity: 56%




Isometric pinch test The test is performed by having the patient pinch the scapulas together posteriorly in retraction. A positive test for scapular muscle weakness is if the patient has burning pain prior to holding this position for 15 to 20 seconds. Unable to find any tests of sensitivity or specificity
Scapular assistance test The scapular assistance test involves assisting the lower trapezius by stabilizing the upper medial border of the scapula and rotating the inferomedial border as the arm is abducted or adducted. The test result is positive, indicating lower trapezius weakness as part of the injury, when it gives relief of symptoms of impingement, clicking, or rotator cuff weakness.


  • Wright et al.




    • Sensitivity: 100



    • Specificity: 33



    • Likelihood ratio: 1.49



Intertester reliability has been studied. In the scapular plane, there was 77% agreement between two examiners, and the kappa coefficient was 0.53. In the sagittal plane, there was 91% agreement between two examiners, and the kappa coefficient was 0.62. The authors concluded that it has acceptable interrater reliability.
Scapular retraction test The test involves manually positioning and stabilizing the entire medial border of the scapula. This test is helpful in two groups of patients. The first group has decreased retraction and apparent muscle weakness. The test result is positive when retesting reveals increased muscle strength with the scapula in the stabilized position. The second group has a positive Jobe relocation test. The test result is positive when scapular retraction decreases the pain or impingement associated with the Jobe relocation test. This indicates that decreased scapular retraction is a component of the overall injury and must be addressed in rehabilitation. A positive scapular retraction test indicates trapezius and rhomboid weakness. We have found no tests assessing the validity, reliability, sensitivity, specificity, positive predictive value, or negative predictive value of this test.


The role of the scapula in normal and abnormal shoulder conditions has been controversial. Kibler and coworkers suggested that changes in scapular position contribute to rotator cuff symptoms, labral tears, and shoulder pain. These conclusions are based on observations that patients with shoulder pathologies often have what appear to be malpositioning of the scapula at rest and abnormal motion of the scapula upon the chest wall with activity. This abnormal scapular motion on the thorax with activity has been called “scapular dyskinesis.” Although there is little doubt that there are scapular dyskinesia patterns, it is unknown whether the patterns are a cause of shoulder pathologies or the result of shoulder pathologies. Therefore, scapular movement issues are typically addressed simultaneously with the painful conditions associated with the scapular motions.


Although measurement of scapular position and movement had become very popular, these concepts have undergone increasing scrutiny. The tests are described below in detail, but the relationships between these findings and the pathophysiology of the clinical findings is being questioned. For example, Kibler and associates proposed that there were four patterns of scapular dyskinesia. Subsequent study found that independent observers could not agree when trying to classify dyskinesia patterns, and the study concluded that agreement was best when the observers merely made a “yes” or “no” assessment of the presence of dyskinesia. Similarly, it was originally suggested that dyskinesia patterns could be associated with specific disease states. This has since been disproven, and although scapular dyskinesia can be associated with a variety of shoulder conditions, it cannot be used reliably as a diagnostic tool for specific shoulder conditions. Consequently, these tests should be used with an understanding of their limitations and clinical applications.


Lateral Scapular Slide Test ( )


Kibler described the lateral scapular slide test (LSST) in identification of subtle ST motion abnormalities as follows Fig. 4.20 :




The first position is with the arms relaxed at the sides. The second is with the hands on the hips with the fingers anterior and the thumb posterior with about 10 degrees of shoulder extension. The third position is with the arms at or below 90 degrees of arm elevation with maximal internal rotation at the glenohumeral joint.


These positions offer a graded challenge to the functioning of the shoulder muscles to stabilize the scapula. The final position presents a challenge to the muscles in the position of most common function at 90 degrees of shoulder elevation …


The reference point on the spine is the nearest spinous process, which is then marked with an X . The measurements from the reference point on the spine to the medial border of the scapula are measured on both sides. In the second position, the new position of the inferomedial border of the scapula is marked, and the reference point on the spine is maintained. The distances once again are calculated on both sides. The same protocol is done for the third position.




Figure 4.20


Lateral scapular slide test. Measurements are made from a reference point (eg, nearest spinous process) to the inferomedial border of the scapula. A, Initial position for examination of the scapula. B, Measurement in the first position with arm relaxed at the side. C, Second position with hands on hips, fingers anterior, and thumb posterior with approximately 10 degrees of shoulder extension. D, Third position with arms at or below 90 degrees of elevation with maximal internal rotation at the glenohumeral joint.


The exact amount of asymmetry that should be considered pathologic is controversial. Kibler defined 1.5 cm of asymmetry as positive for ST motion abnormality. Odom and coworkers reported 1 cm of asymmetry as being positive when correlated with patients who did or did not have shoulder pathologies. The sensitivities and specificities of this test for pathologic conditions were low regardless of the position measured. Odom and coworkers concluded that “the LSST should not be used to identify people with [or] without shoulder dysfunction.”


Isometric Pinch Test


In Kibler’s 1998 paper, “the role of the scapula in athletic shoulder function” is described by a provocative maneuver for evaluating scapular muscular strength. Kibler writes:




A good provocative maneuver to evaluate scapular muscle strength is to do an isometric pinch of the scapulae in retraction. Scapular muscle weakness can be noted as a burning pain in less than 15 seconds. Normally, the scapula can be held in this position for 15 to 20 seconds with the patient having no burning pain or muscle weakness.



No independent studies have validated this test or examined its clinical utility. A similar test is the costoclavicular maneuver for making the diagnosis of thoracic outlet syndrome.


Scapular Assistance Test ( )


Another test for the strength of the scapular stabilizers is the scapular assistance test ( Fig. 4.21 ) described by Kibler and McMullen in 2003. They described the test as follows:




The scapular assistance test evaluates scapular and acromial involvement in subacromial impingement. In a patient with impingement symptoms with forward elevation or abduction, assistance for scapular elevation is provided by manually stabilizing the scapula and rotating the inferior border of the scapula as the arm moves. This procedure simulates the force-couple activity of the serratus anterior and lower trapezius muscles. Elimination or modification of the impingement symptoms indicates that these muscles should be a major focus in rehabilitation.




Figure 4.21


The scapular assistance test is designed to determine if stabilizing the scapula improves shoulder pain. The examiner stabilizes the scapula and elevates the arm. The arrow shows the direction of the examiner’s force.


There has been no independent verification of this study, and its clinical usefulness has not been adequately studied.


Scapular Retraction Test


The scapular retraction test was described by Kibler and associates to distinguish a scapular cause of weakness of the supraspinatus. After initial standard supraspinatus testing (Jobe test), the medial border of the scapula is stabilized by the examiner, and muscle testing is repeated. The test is considered positive if supraspinatus strength increases after stabilization of the scapula. We have found no reports assessing the sensitivity, specificity, PPV, or NPV of this test.


Tests of the Biceps Tendon


See Table 4.5 .



Table 4.5

Tests of the Biceps Tendon
























Test Description Reliability/Validity
Yergason’s test The elbow is flexed to 90 degrees with the forearm pronated, and the examiner holds the patient’s wrist to resist supination and then directs that active supination be made against the resistance; pain, very definitely localized in the bicipital groove, indicates a condition of wear and tear of the long head of the biceps.


  • Calis et al.




    • For diagnosis of subacromial impingement (not evaluating the biceps tendon) using MRI and Neer injection test as the gold standards:




      • Sensitivity: 37%



      • Specificity: 86%



Speed’s test Have the patient flex the shoulder (elevate it anteriorly) against resistance while the elbow is extended and the forearm supinated. The test result is considered positive when pain is localized to the bicipital groove.


  • Bennett




    • Sensitivity: 90%



    • Specificity: 13.8%




  • Calis et al.




    • For subacromial impingement:




      • Sensitivity: 68.5%



      • Specificity: 55.5%



Lift-off test for partial tears of the biceps tendon The hand of the affected arm is placed on the back at the midlumbar region, and the patient is asked to rotate the arm internally and lift the hand posteriorly off the back. A positive test result is when the patient cannot lift the hand off of the back.


  • Gill et al.




    • Sensitivity: 28%



    • Specificity: 89%



    • PPV: 15%



    • NPV: 95%



    • Likelihood ratio: 2.61



    • Accuracy: 85%


Ludington test The patient is asked to put hands on the head with palms down and to contract the biceps muscle. The test result is positive if there is a visible deformity or if the biceps tendon cannot be felt proximally in the arm. There are no reported studies assessing the sensitivity, specificity, PPV, or NPV of this maneuver.

MRI, magnetic resonance imaging; NPV, negative predictive value; PPV, positive predictive value.


Physical examination tests of the biceps tendon present challenges to the clinician. There are several reasons for this. First, the biceps tendon is deep in the joint where it cannot be palpated. Also, even the extra-articular part of the tendon in the bicipital groove is difficult to palpate because other structures (namely the rotator cuff tendons) attach near the bicipital groove. Second, a click or a catch in the shoulder cannot be assumed to be caused by the biceps tendon. One study found that only 5% of patients with superior labral tears have a click, but 5% of a control group also had a click.


Ludington Test


In 1923, Nelson Ludington described a test for diagnosing rupture of the long head of the biceps. Ludington asked the patient to put his or her hands on the head with the palm down and to contract the biceps muscle ( Fig. 4.22 ). The test result was positive if there was a visible deformity of the biceps (Popeye deformity) or if the biceps tendon could not be felt proximally in the arm. This test has never been studied clinically, but palpation of the long head of the biceps tendon is not typically reliable in the proximal arm. Also, in most patients with a torn biceps tendon, a bulge is seen simply by asking the patient to contract the biceps muscle with the arm at the side.




Figure 4.22


The Ludington test was designed to compare the biceps muscle shape side to side. The arrow shows the Popeye deformity.


There are no reported studies assessing the sensitivity, specificity, PPV, or NPV of this maneuver.


Yergason’s Test


Robert Yergason originally described his “supination sign” for evaluating tendonitis of the biceps tendon in 1931. He described the test as follows ( Fig. 4.23 ):




If the elbow is flexed to 90 degrees, the forearm being pronated; and the examining surgeon holds the patient’s wrist so as to resist supination, and then directs that active supination be made against his resistance; pain, very definitely localized in the bicipital groove, indicates a condition of wear and tear of the long head of the biceps… ( )




Figure 4.23


Yergason’s test is performed by the examiner resisting forearm supination by the patient with the elbow bent. The arrow shows the direction of the examiner’s force.


Calis and associates found the Yergason’s test to have a sensitivity of 37% and a specificity of 86.1% for diagnosis of subacromial impingement using MRI and Neer injection test as the gold standards. The authors described the test for a disorder of the long head of the biceps tendon but did not specify how this related to the diagnosis of biceps disease or conditions.


Speed’s Test


The earliest reference to this study in the literature was by Crenshaw and Kilgore on “the surgical treatment of bicipital tenosynovitis” in 1996. They cite a personal communication with Speed in 1952 and describe the test as follows ( Fig. 4.24 ):




[Have] the patient flex his shoulder [elevate it anteriorly] against resistance while the elbow is extended and the forearm supinated. The test is considered positive when pain is localized to the bicipital groove ( ).




Figure 4.24


Speed’s test is performed by the patient resisting a downward force by the examiner (arrow). A positive test result is pain in the biceps area of the shoulder.


Several studies have shown that Speed’s test does not actually help the clinician in making the diagnosis of biceps tendon disorders. Bennett found Speed’s test to have a specificity of 13.8% and a sensitivity of 90% for biceps tendon disorders. Gill and coworkers found that Speed’s test had a sensitivity of 50%, specificity of 67%, PPV of 8%, NPV of 96%, and likelihood ratio of 1.51 for detecting partial tears of the biceps tendon. Calis and associates noted the Speed’s test to have a sensitivity of 68.5% and a specificity of 55.5%. Burkhart and others evaluated Speed’s test for labral pathology. They found that it had a sensitivity of 100% and a specificity of 70% for anterior labral lesions and a sensitivity of 29% and a specificity of 11% for posterior labral lesions. The combined sensitivity and specificity for both lesions were 78% and 37%, respectively.


Tests of Rotator Cuff Disease


See Table 4.6 .



Table 4.6

Tests of Rotator Cuff Disease




























Test Description Reliability/Validity
Neer sign Passive elevation of the arm in flexion with the arm in internal rotation while stabilizing the scapula from the back should result in pain into the deltoid region. Typically, pain occurs around 120 degrees of flexion.


  • Calis et al.




    • Sensitivity: 88.7%



    • Specificity: 30.5%




  • MacDonald et al.




    • For assessing subacromial bursitis:




      • Sensitivity: 75%



      • Specificity: 47.5%




    • For detecting RC pathology:




      • Sensitivity: 83.3%



      • Specificity: 50.8%





  • Park et al.




    • For assessing subacromial bursitis:




      • Sensitivity: 85.7%



      • Specificity: 49.2%



      • PPV: 20.9%



      • NPV: 95.7%



      • +LR: 1.69



      • Accuracy: 54.2




    • For partial RC tears:




      • Sensitivity: 75.4%



      • Specificity: 47.5%



      • PPV: 18.1%



      • NPV: 92.6%



      • +LR: 1.44



      • Accuracy: 51.3




    • For full-thickness RC tears:




      • Sensitivity: 59.3%



      • Specificity: 47.2%



      • PPV: 41.3%



      • NPV: 64.9%



      • +LR: 1.12



      • Accuracy: 51.8%



Neer subacromial injection test Neer sign pain may be temporarily stopped by instilling 1% lidocaine into the bursa. There are no studies that validate the Neer test.
Hawkins test This involves forward flexing the humerus to 90 degrees and internally rotating. Pain should radiate into the deltoid region.


  • MacDonald et al.




    • For assessing subacromial bursitis:




      • Sensitivity: 91.7%



      • Specificity: 44.3%




    • For RC pathology:




      • Sensitivity: 87.5%



      • Specificity: 42.6%





  • Calis et al.




    • Sensitivity: 92.1%



    • Specificity: 25%




  • MacDonald et al.




    • When Neer and Hawkins tests were both positive for detecting bursitis:




      • Sensitivity: 70.8%



      • Specificity: 50.8%




    • For detecting RC pathology:




      • Sensitivity: 83.3%



      • Specificity: 55.7%




    • If only one of the two tests was positive, for detecting bursitis:




      • Sensitivity: 95.8%



      • Specificity: 41%




    • For detecting RC pathology:




      • Sensitivity: 87.5%



      • Specificity: 37.7%



Painful arc sign The patient is asked to actively abduct the shoulder. In a positive test result, the patient will experience pain from approximately 70 to 120 degrees, and pain will diminish after that level of elevation. The pain is typically into the deltoid area and sometimes worsens when bringing the arm down from an elevated position.


  • Park et al.




    • For subacromial bursitis:




      • Sensitivity: 70.6%



      • Specificity: 46.9%



      • PPV: 12.3%



      • NPV: 93.8%



      • LR: 1.33



      • Accuracy: 49.2%




    • For partial RC tears:




      • Sensitivity: 67.4%



      • Specificity: 47%



      • PPV: 14.9%



      • NPV: 91.3%



      • LR: 1.27



      • Accuracy: 49.4%




    • For full-thickness RC tears:




      • Sensitivity: 75.8%



      • Specificity: 61.8%



      • PPV: 61%



      • NPV: 76.4%



      • LR: 1.98



      • Accuracy: 68%



Yocum’s test The patient is asked to place the hand on his or her other shoulder and to raise the elbow without elevating the shoulder. This test is positive when it elicits the pain usually experienced by the patient.


  • Naredo et al.




    • Yocum’s test in combination with Hawkins’ and Neer’s test:




      • Sensitivity: 65%



      • Specificity: 72.7%





  • Silva et al.




    • Sensitivity: 79%



    • Specificity: 79%



    • LR: 1.32


Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Jul 23, 2019 | Posted by in MUSCULOSKELETAL MEDICINE | Comments Off on Physical Examination of the Shoulder

Full access? Get Clinical Tree

Get Clinical Tree app for offline access