Chapter 7 The shoulder girdle

The use of plain imaging is still an essential starting point when investigating shoulder trauma,¹ and basic diagnostic errors may occur if other imaging modalities are used alone, without the use of conventional plain radiography.² Guidelines still suggest that plain radiography is indicated for fractures, dislocations, shoulder instability and calcific tendonitis, with magnetic resonance imaging (MRI), ultrasound and computed tomography (CT) suggested for soft tissue injuries and ‘more complex cases’. More specifically, ultrasound is recommended for rotator cuff injuries.²

Ultrasound can be used to assess disorders such as defects in the long head of the biceps tendon.² Subacromial and acromioclavicular joint impingement are dynamic processes and these can also be studied during ultrasound examination.

CT may be used in preoperative assessment of shoulder injuries, so that fractures are not underestimated (as can be the case in some instances with plain X-ray images) and 3D reconstruction is often used to fully demonstrate complex fractures and assist in surgical planning.¹

MRI has become an increasingly important technique for evaluating rotator cuff disorders and joint instability,^1,² its effectiveness being due to high contrast sensitivity and multiplanar imaging capabilities; therefore, diagnosis and appropriate management of the complex shoulder joint is established with greater confidence.

When imaging this region with plain radiography, radiation protection of the eyes and thyroid is an important consideration: the patient must always have their head turned away from the primary beam during exposure.

This area of high subject contrast has implications for overexposure of some structures involved in the joint. This is especially true of the acromioclavicular joint, which is often lacking in detail due to overexposure, whereas details of denser structures of the region (e.g. the humeral head or glenoid) are adequately demonstrated. Repeat examinations are often required as a result, and can be avoided in the first instance by using a wedge filter placed between the image receptor (IR) and the upper shoulder. The most effective type of filter for this is rubberised and boomerang shaped and can therefore sit comfortably and safely around and behind the upper shoulder. Use of a relatively high kVp and lower mAs can offer a solution in the absence of a filter, but the contrast of these images is somewhat reduced compared with those produced with a filter. Patients with very dense muscle (e.g. body builders and rugby players) will certainly need effective beam penetration.

Indications

Arthropathy

Erosions are a relatively late feature in patients with rheumatoid arthritis and the shoulder should only be examined by plain imaging if that joint is specifically affected. In patients with suspected osteoarthritis, X-ray is not indicated initially unless intervention is likely.

Fracture

This mostly affects the clavicle, humeral surgical neck, tuberosities of the humerus and scapula; fracture of the scapula is relatively uncommon, accounting for only 3–5% of shoulder injuries.

Fractures of the surgical neck of the humerus and the tuberosities have often been classified using Neer’s method,³ which considers the status and degree of displacement of the articular segment of the head of the humerus, the surgical neck of the humerus and the greater and lesser tuberosities. The reliability of such classification systems has been questioned and alternative classification methods suggested;⁴ however, it must be mentioned that new methods, however reliable, need to be widely accepted so that they can be considered rigorous.

Dislocation

The shoulder joint is the most commonly dislocated joint in the human body,⁵ with anterior dislocation most common; only up to 5% of dislocations occur posteriorly,⁶ and an estimated 60–80% of these are missed on initial examination. As many as 50% of these uncommon dislocations can often be missed in A&E, highlighting the importance of an additional projection that can identify posterior dislocations.⁷ Subluxation of the acromioclavicular joint can also occur.

Anteroposterior (AP) shoulder (Fig. 7.1A,B)

This projection can be performed in the erect position, either standing or seated, depending on the patient’s condition and ability. When examining a patient on a trolley, care should be taken to ensure either that the patient is in the fully erect position or the beam is accurately angled to compensate for any tilt on the trolley back rest; this will ensure that the central ray remains at 90° to the IR.

Figure 7.1 AP shoulder.

The AP projection does not demonstrate the glenohumeral joint space clearly and orthopaedic departments may request either a ‘True AP’ or ‘Grashey AP’³ instead of, or to complement, the AP. This projection uses the same position and centring point as the AP described here but with an obliquity of the patient at 45° instead of 20°, to open the glenohumeral joint. The view shows the glenohumeral joint tangentially, but published work varies in assessment of its effectiveness in demonstration of direction of dislocation.⁸^–¹⁰

Common errors	Possible reasons
Inferior end of the scapula not included on the image	The IR is often positioned in the ‘landscape’ position; putting it in the ‘portrait’ position will usually prevent this
Foreshortening of the clavicle	The patient is rotated too much towards the side under examination
The acromioclavicular joint is over-penetrated	This is due to the difference in subject contrast in this area; the use of a wedge filter will prevent this

Axillary/axial projections of the shoulder

Evaluation of the shoulder joint, particularly for follow-up orthopaedic assessment, often requires an axillary projection to offer an image at 90° to the AP. Success of this projection will depend on the patient’s condition and cooperation. Two methods are described here: method 1 is often difficult to implement or inappropriate, particularly in trauma, owing to the extent to which the arm must be abducted. Method 2 is the method of choice for a patient with restricted movement of the humerus as there is more scope for adaptation to suit the patient’s condition. Method 2 is sometimes referred to as the Lawrence axillary.¹¹

Method 1: superoinferior shoulder (Fig. 7.2A,B)

Positioning

• The IR is horizontal

• The patient sits with the side under examination next to and slightly away from the table

• For radiation protection purposes the legs are placed so they are not under the table and a lead rubber apron is worn around the waist

• The arm is abducted fully and the patient leans laterally over the IR; the hand is internally rotated and pronated. The axilla is positioned over the IR in a position that will ensure inclusion of the relevant anatomy, and with the axilla as close to it as possible

• The patient’s head and neck are abducted away from the shoulder under examination as far as possible to clear them from the area of interest and reduce the radiation dose to these areas

Figure 7.2 Superoinferior shoulder.

Beam direction and FRD

Vertical at 90° to the IR

100 cm FRD

Centring point

To the superior aspect over the middle of the head of the humerus

Collimation

Head and proximal third of humerus, glenoid cavity, acromion, coracoid process, surrounding soft tissues

Criteria for assessing image quality

• Head and proximal end of humerus, glenoid fossa, lateral end of clavicle, acromion and coracoid process are demonstrated

• Head of humerus appears above the glenoid ‘like a golf ball on a tee’⁷

• Greater tuberosity should be seen in profile anteriorly

• Acromion and lateral end of clavicle are superimposed on the superoposterior aspect of the head of humerus

• Coracoid process is demonstrated anterior to the head of humerus

• Sharp image demonstrating the soft tissue margins, bony cortex and trabeculae of the head of the humerus with adequate image density to demonstrate the bony detail of the humerus in contrast to the glenohumeral joint, acromion and clavicle

Common errors	Possible reasons
The glenohumeral joint is not demonstrated within the boundaries of the IR	The patient may not be stretching across the IR sufficiently. If the patient is capable of leaning further, try lowering the table-top to enable the patient to flex more at the waist
Magnification and unsharpness of the resulting image, probably accompanied by foreshortening of humeral head	The axilla is not in close enough contact with IR and the humerus may not be fully abducted, causing its shaft to lie at an angle with the IR. Try using a pad to raise the IR or consider increasing the FRD to compensate for the large ORD

Method 2: inferosuperior shoulder; ‘Lawrence axillary projection’ (Fig. 7.3)

Positioning

• The patient lies supine on the table

• A small radiolucent pad is placed beneath the shoulder to raise it slightly

• The head and neck are abducted as much as possible away from the side under examination to clear them from the area of interest and reduce radiation dose to these areas

• The IR is supported in the erect position, its tube side against the superior aspect of the head of the humerus and in contact with the neck

• The arm is abducted to 90° or as far as the patient’s condition permits and the hand is supinated (although some internal rotation of the forearm is acceptable; supination acts mainly to help the patient maintain the correct relationship of the humerus to the IR)

• A lead rubber apron is placed over the patient’s chest and abdomen for radiation protection

Figure 7.3 Inferosuperior shoulder.

Modern tube housings are usually too bulky to allow tube centring for this positioning. An alternative, modified inferosuperior is suggested, as follows:

This technique can be achieved with as little as 30° arm abduction,³ but the tube needs to be brought in as close to the patient’s body as possible. By lying the patient in a slightly diagonal position across the length of the table-top or trolley, access to the axilla is achievable (Fig. 7.4). Positioning the patient thus, diagonally across the table-top, requires consideration for the safety of the patient; this is directly related to table width and should only be considered in the relatively cooperative patient.

Figure 7.4 Modified inferosuperior shoulder.

Beam direction and FRD (inferosuperior and modified inferosuperior projections)

Horizontal at 90° to the IR and coincident with the glenohumeral joint

100 cm FRD

The central ray must be at 90° to the IR and requires careful positioning to prevent a distorted image. To eliminate distortion, align the central ray with the patient first to ensure it is parallel to the glenohumeral joint, i.e. through the axilla, then position the IR until perpendicular to the central ray. This is suggested for both the inferosuperior and modified inferosuperior projections.

Centring point

Through the axilla

Collimation

Head and proximal third of humerus, glenoid cavity, acromion, coracoid process, surrounding soft tissues

Criteria for assessing image quality

• Head and proximal end of humerus, glenoid fossa, lateral end of clavicle, acromion and coracoid process should all be demonstrated

• Glenohumeral joint should be demonstrated

• Lesser tuberosity of humerus should be seen in profile

• Acromioclavicular joint will be superimposed on humerus

• Sharp image demonstrating the soft tissue margins, bony cortex and trabeculae of the head of humerus with adequate image density to demonstrate the bony detail of humerus in contrast to the glenohumeral joint

Common error	Possible reason
The glenohumeral joint is not demonstrated within the boundaries of the IR	The head and neck may not be sufficiently abducted away from the side under examination to enable the IR to be positioned correctly. Always ensure the IR is closely tucked into the neck

The inferosuperior projections can be adapted to demonstrate the classic Hill–Sachs compression fracture, seen in patients who have recurrent anterior dislocation of the shoulder.¹ This adaptation involves maximum external rotation of the arm, with the patient aiming to press the thumb down towards the table or trolley top. Unfortunately, this manoeuvre can be difficult for patients to achieve and the AP shoulder with maximum internal rotation can also demonstrate this lesion adequately.² In order to achieve the correct amount of rotation for this, the arm is medially rotated and flexed at the elbow; the dorsum of the hand is then rested on the waist. Yet another technique that can show Hill–Sachs lesions is the Stryker notch view, where the palm of the hand is placed on top of the head with the fingers toward the back of the head and the long axis of the humerus parallel to the median sagittal plane (MSP); a 10° cranial angle is centred over the coracoid process for this projection.¹²

Bankart lesions are also recognised as an effect of recurrent anterior dislocation ¹³ and are best seen on the true superoinferior view; clearly, superoinferior is recognised as difficult on the traumatised patient but feasible on patients with recovered range of shoulder movement following treatment.

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