The standard AP view of the shoulder,
in conjunction with the findings on physical examination, is usually
sufficient to diagnose an anterior dislocation (Figure 2A).
Although an AP radiograph does not allow direct determination
of whether the dislocated humeral head is anterior or posterior
to the glenoid fossa, anterior dislocation can be surmised by making
two observations. First, with an anterior dislocation, displacement
of the humeral head is much greater than is possible with a posterior
dislocation. The humeral head is usually in a subcoracoid position,
although subglenoid and subclavicular dislocations occasionally
occur. With a posterior dislocation, displacement is limited by
the muscles of the rotator cuff. Second, with an anterior dislocation
the humerus is fixed in external rotation (greater tuberosity is
lateral) and an internal rotation view cannot be obtained. Internal
rotation of the shoulder is prevented by impaction of the humeral
head on the anterior rim of the glenoid fossa.
To directly demonstrate the proximal humerus dislocation, an
additional view such as a scapular Yview or a axillary
view can be obtained (Figure 2B and Patient 6, Figures 8 and 11, in Posterior Shoulder Dislocation). Although these views are frequently
obtained in patients with anterior dislocations, they are generally
not essential to make that diagnosis. Occasionally, these views
will reveal a scapular fracture that was not evident on the AP view.
Because anterior shoulder dislocations are usually obvious on
physical examination, radiographs are generally not necessary to
make the diagnosis. Instead, the principal role of radiography is
to detect associated fractures, which occur in 8–25% of
shoulder dislocations (Perron et al. 2003, Emond et al. 2004). (This
does not include Hill-Sachs deformities, which occur in up to 25% of dislocations.)
However, in most cases, the fracture does not alter emergency management.
The most common fracture associated with an anterior shoulder
dislocation is an avulsion of the greater
tuberosity of the humerus, occurring in 5–19% of
cases (Figure 3). Avulsion occurs because the supraspinatus
tendon is attached to the greater tuberosity (Figure 4). This
fracture does not alter the technique of reduction because when
the shoulder is reduced, the avulsed fragment usually returns to
its proper position (Figure 3B). If the fracture remains displaced
by 1 cm or more, surgical screw fixation is indicated, particularly
in younger patients.
Proximal humerus fractures associated with shoulder dislocations
occur more commonly in older individuals. Fractures are seen in
as many as 40% of shoulder dislocations in patients over
age 50 years. The radiographs therefore must be carefully inspected
for evidence of a fracture.
In Patient 7, a greater tuberosity fracture was recognized, in
addition to the anterior dislocation. However, a non-displaced fracture
of the humeral neck was not appreciated. Standard closed reduction
using traction/counter-traction was performed. Post-reduction
radiographs revealed that the attempted reduction caused complete
displacement of the humeral head from the humeral shaft (Figure
Patient 7—Post-reduction radiographs
The AP view (A) and Y view (B) revealed complete displacement of
the humeral head (arrows) due to an
unsuspected humeral neck fracture. The humeral shaft was “reduced,” but
the humeral head remained dislocated.
(C) Postoperative view showing hemi-arthroplasty
of the proximal humerus.
Because of the risk of avascular necrosis as well as the potential
difficulty attaining internal fixation using plates and screws in
this elderly woman, the patient was treated with a hemi-arthroplasty
Review of the initial AP radiograph reveals a subtle fracture
of the humeral neck. It appears as trabecular impaction and cortical
disruption at the humeral neck (Figure 6). Had this been recognized,
closed reduction should not have been attempted.
Patient 7—Initial AP view
Initial AP view showing an anterior dislocation, avulsion fracture
of the greater tuberosity (arrowheads)
and a subtle fracture through the humeral neck (arrows).
(The metallic rectangle is a clip for the shoulder sling.)
Humeral neck fracture/dislocations are less common than
greater tuberosity avulsion fractures, occurring in 2.6–4% of
cases (Perron et al. 2003, Emond et al. 2004). However, they are
the second most common fracture associated with anterior shoulder
dislocations and are the most important reason to obtain radiographs
in patients with clinically evident anterior shoulder dislocations because
management is substantially altered.
The humeral head cannot be reduced by applying traction on the
humeral shaft. Attempted closed reduction can cause further separation
of the fracture fragments, increasing the risk of avascular necrosis
of the humeral head (Figure 7).
A 40-year-old man with an anterior dislocation and
displaced humeral neck fracture (A and B).
(C and D)
During attempted closed reduction, the humeral shaft was re-aligned
with the glenoid fossa, but the humeral head remained dislocated.
Closed reduction should not have been attempted because it was ineffective
and further displaced the fracture.
(E) Open reduction was necessary.
The humeral neck fracture was treated with plate and screws fixation.
A nondisplaced humeral neck fracture can be difficult to visualize
but should be specifically sought, particularly in older individuals
with shoulder dislocations.
When a humeral neck fracture is found or suspected, closed reduction
can still be attempted, although this should be done either in the
operating room under general anesthesia with fluoroscopic guidance
(Hersche and Gerber 1994), or possibly in the ED, after consultation
with an orthopedic surgeon, using adequate sedation, analgesia and
gentle traction while manipulating the displaced humeral head. If
unsuccessful, open reduction should be performed. If an inapparent humeral
neck fracture is displaced during attempted reduction, immediate
open reduction and internal fixation has been suggested, particularly
for younger patients, in the hope of decreasing the incidence of
later avascular necrosis of the humeral head (Ferkel et al. 1984).
An injury that should not be mistakenly interpreted as a fracture-dislocation
is a pseudo-dislocation (Figure 8).
In a patient with a proximal humerus fracture, a hemarthrosis can
distend the joint capsule and displace the humeral head from the
glenoid fossa. On the AP radiograph, this displacement can mimic
dislocation of the glenohumeral joint. A Y-view or axillary view
may be needed to show that the humeral head is still in alignment
within the glenoid fossa. Traction on the arm in an attempt to reduce
this injury should not be attempted because it will further displace the
fracture increasing the risk of avascular necrosis of the humeral
Pseudo-dislocation of the shoulder.
(A) The AP view shows a minimally
displaced humeral neck fracture with inferior displacement of the
humeral head that mimics a dislocation. This is due to a hemarthrosis
distending the joint capsule and is known as a pseudo-dislocation.
(B) On the Y-view, the humeral head
is inferiorly and anteriorly displaced. Although the Y-view can usually
distinguish a dislocation, in this patient, glenohumeral subluxation
mimics an anterior shoulder dislocation. This illustrates the limitations
of the Y view in this situation.
(C) An axillary view confirms that
there is no dislocation.
for Radiography in Patients with Anterior Shoulder Dislocation
Standard management of a patient with suspected anterior shoulder
dislocation is to obtain both pre- and postreduction radiographs.
However, because both an anterior dislocation and its successful
reduction are usually obvious on clinical examination, the role
of radiography is controversial. The two principal roles of radiography
are (1) to establish the diagnosis of dislocation when it is uncertain
and (2) to detect fractures that have occurred in association with
Pre-reduction radiographs could
theoretically be omitted if the diagnosis was certain based on clinical
examination. However, when the traumatic force causing the dislocation
could also cause a fracture, prereduction radiographs should be
obtained whenever possible. In settings where radiography is not
available, such as when the injury occurs on an athletic field,
reduction could be attempted without radiography to avoid a prolonged
delay in treatment (Shuster 1999). One instance when radiography
may not be necessary is in a patient with recurrent shoulder dislocations
in whom the dislocation occurred with minimal force such as by simply
abducting the arm. In this case, there is clearly insufficient force
to cause a fracture.
Emond et al. (2004) identified three factors predictive of a
fracture in association with a shoulder dislocation: age 40 years
or older, first episode of dislocation, and dangerous mechanism
of injury (fall greater than one flight of stairs, fight/assault
or motor vehicle crash). Absence of all three features was 98% sensitive
at excluding an associated fracture and could potentially be used
to omit pre-reduction radiographs, although this has not yet been
The necessity of post-reduction radiographs has
been questioned because reduction is usually clinically evident
by noting the change in shoulder contour and increased range of
motion. Based on a prospective series of 104 patients, Hendey (2000)
suggested that in patients without a fracture, when clinicians were
confident of the success of reduction (93% of 85 cases),
post-reduction radiographs could be omitted because the clinicians
were correct in all cases. When the clinicians were not confident
of the reduction (6 cases) or there was a concomitant fracture (10
cases), post-reduction radiographs were necessary.
Concern that post-reduction radiographs should be obtained because
a fracture could be caused by the force of manipulation is implausible
given current nontraumatic techniques of reduction.
Hendey (2000) proposed an algorithm defining
two circumstances in which radiography could potentially be omitted.
First, pre-reduction radiographs could
be omitted in patients with recurrent dislocation when mechanism
of injury was atraumatic (28% of patients). Second, post-reduction radiographs could be
omitted when there were no associated fractures seen on the prereduction radiographs
and the clinicians were confident of the reduction. In his series,
this would reduce the use of radiography by 51%. A small
prospective series found that the algorithm reduced radiograph utilization
by 47%, without missing any clinically significant fractures
or persistent dislocations (Hendey et al. 2004).
Post-reduction radiographs may show fractures that were not seen
on the prereduction radiographs, although these fractures generally
do not alter acute management, such as Hill-Sachs deformities (Figure
2) (Hendey and Kinlaw 1996, Harvey et al. 1992). Nonetheless, the
fractures may have long term implications and radiographs would
likely be desired by the orthopedist during follow-up of the patient.
In addition, if a subtle humeral neck fracture was unrecognized
prior to reduction, as in this patient, failure to obtain post-reduction
radiographs could result in significant morbidity.