SHOULDER INSTABILITY

The history can raise suspicion of shoulder instability.
A patient’s age, occupation, employment status, activity level, hand
dominance, and other medical problems are helpful in classifying the
nature of the instability. Certain circumstances, such as work-related
injuries or those with associated litigation, might affect patients’
perceptions of their disabilities as well as their expectations for
recovery and compliance with treatment (180).

Seek out specific complaints. For example, a patient may indicate that the arm “went dead” during a football tackle (158).
This symptom is likely associated with a traumatic subluxation of the
joint and the presence of a Bankart lesion. Another example is a
patient who gives a

history
of having had a seizure and then awakening with the sensation of the
joint “out of place.” This situation suggests a posterior dislocation
as the result of unchecked internal rotation forces overcoming the
weaker external rotators of the shoulder.

If pain is a major component of a patient’s complaint,
which activities cause the pain? Question athletes as to the type of
sport, the position played, and the duration, frequency, and level of
involvement. For example, throwers, swimmers, or tennis players may
develop excessive capsular laxity as a result of subjecting their
shoulders to repetitive microtrauma (83,125).
The result is recurrent shoulder subluxation, which can create a sense
of “looseness” or “slipping” with activity. In addition, as previously
stated, secondary “nonoutlet” impingement can occur as anterior
instability decreases the subacromial space, causing pain with overhead
activities. Further, these same subluxations can create damage to the
superior labral area, creating a SLAP lesion. Posterior subluxation has
also been documented in athletes who require repetitive arm motion in
front of the body, such as offensive-line football players and
volleyball, softball, and baseball players (48,156,178). Throwing sports in particular lead to complaints of posterior
shoulder pain during the follow-through phase, which has been
associated with posterior subluxation and resultant microtrauma to the
posterior capsulolabral complex.

A systematic evaluation includes observation for
abnormal motion patterns and atrophy, palpation to localize painful
areas, assessment of both active and passive range of motion,
measurement of strength of the rotator cuff muscles, neurovascular
evaluation, and finally provocative maneuvers for instability. Examine
the opposite shoulder for comparison.

In evaluating shoulder motion, carefully document any
scapulothoracic substitution for glenohumeral motion, scapular winging,
and other abnormal motion patterns. Atrophy of the spinatus muscles may
indicate a longstanding associated rotator cuff tear or injury to the
suprascapular nerve. Similarly, atrophy of the deltoid may indicate an
axillary nerve injury.

Assess patients for findings of generalized ligamentous
laxity, including the ability to hyperextend the elbows more than 10°,
apply the thumb to the forearm, hyperextend the metacarpophalangeal
joints more than 90°, or touch the palm of each hand to the floor while
keeping the knees extended (Fig. 80.8). While
there is no direct relationship between generalized ligamentous laxity
and shoulder instability, there is some association between hyperlaxity
and glenohumeral capsular development.

A patient with an acute, unreduced anterior dislocation holds the painful arm in slight abduction and internal rotation (2,167).
Before attempting any reduction maneuvers, perform a careful
neurovascular examination to rule out brachial plexus and, more
specifically, axillary nerve injuries (38,57,152).
The latter condition may sometimes escape detection, as decreased
sensation over the lateral deltoid is not always present with an injury
to this nerve. In older (>60 years) individuals or younger ones who
have sustained a severe trauma, be aware of the possibility of an
associated fracture of the humerus (62). Proper radiographic imaging is particularly important before attempting closed reduction in such cases.

Many methods of closed reduction apply in the case of an
acute shoulder dislocation. Perform all maneuvers as a gradual and
gentle technique with appropriate analgesia to ensure muscle
relaxation. Intravenous analgesia and intraarticular injection of a
regional anesthetic are both successful, appropriate methods of
anesthesia. A method of gentle traction in line with the arm using
countertraction is usually successful.

Pay careful attention to a patient who has an
unrecognized chronic (fixed) dislocation. The dislocation is most often
posterior, although it may be anterior, and typically occurs in
patients who are poor historians because of either alcohol use or
dementia (64). In a patient with a fixed
posterior subluxation, there is limitation of external rotation
compared with the opposite shoulder. In addition, there is flattening
of the anterior aspect of the shoulder with an associated prominence of
the coracoid process and possibly some prominence and rounding of the
posterior aspect of the shoulder. The application of excessive force in
attempting a closed reduction in such a patient risks neurovascular
injury or fracture.

Most patients who have instability are first seen by an
orthopaedic surgeon in the office setting. They may have had a
documented episode of instability or an injury with pain but no true
sense of shoulder instability. After a careful neurovascular
examination, it is important to assess active and passive range of
motion. A discrepancy between active and passive motion may indicate
either an associated rotator cuff tear or a nerve injury.

It is particularly important to identify a subscapularis
tear in the setting of shoulder instability, a condition that is
frequently missed (51,165).
Patients with such tears have passively increased external rotation
with the arm adducted at the side, as well as associated apprehension
in this position. Strength assessment is also important. Significant
external rotation weakness may indicate a rotator cuff tear.

Similarly, internal rotation weakness may be due to a
subscapularis tear. In this situation, the patient has an associated
lift-off sign (Fig. 80.9). If the patient lacks
adequate internal rotation to perform this test, perform the
belly-press maneuver instead to determine the patient’s ability to pull
the forearm in a posterior direction toward the mid abdomen while
maintaining the flexed elbow anterior

to
this point. If the elbow remains posterior to the anterior aspect of
the mid abdomen, there is likely a subscapularis tendon tear.

Various provocative maneuvers and drawer tests have been
described to assess symptomatic humeral head translation and the
direction of this movement. Neer and Foster (107)
originally described the apprehension test for anterior instability.
With the patient seated or standing, place the symptomatic shoulder
into a position of 90° of abduction and maximum external rotation. Then
move the arm so that it is posterior to the plane of the scapula, and
apply an anterior force to the humeral head. The patient’s withdrawing
from the examiner or complaining about a sense of shoulder instability
demonstrates apprehension (Fig. 80.10A).

The complaint of pain is not specific for instability
and may in fact be present with other conditions such as arthritis or
rotator cuff disease. Kvitne and Jobe (83)
proposed a modification of this maneuver to increase its specificity
for anterior instability. Place the patient in a supine position, and
perform the apprehension test as described above. Ask whether the
patient has a sense of instability or simply pain. Place posterior
pressure on the humerus, and ask whether this pressure relieves the
sense of apprehension or pain (Fig. 80.10B).
This “relocation maneuver” increases the specificity of the diagnosis
of instability if the patient reports decreased apprehension. If this
maneuver simply reduces pain, it is not diagnostic of instability and
may be associated with a variety of other diagnoses, including a SLAP
lesion or impingement syndrome (150).

One of us (JJPW) has found that an additional
modification to this test increases both its sensitivity and
specificity for the diagnosis of anterior shoulder subluxation. Perform
the apprehension and relocation tests as described above. If the
patient reports pain and not true apprehension, inject the subacromial
space with 10 ml of lidocaine 1% (Xylocaine). After 10 minutes, repeat
the examination. If the pain is due to rotator cuff disease, the
injection will eliminate it. If, however, the pain is due to labral
injury, it will persist after a subacromial injection (166).

Inconsistencies with the apprehension test led Gerber and Ganz (50)
to develop the anterior and posterior drawer test to assess the
shoulder for excessive translation compared with the contralateral
side. Others have found merit in this method of examination and have
developed grading scales for the degree of shoulder laxity (3,30,65).
These tests may offer some insight into the degree and direction of
instability. If one assesses laxity of the shoulder in an office
setting, it is important to determine whether translation of the
humeral head is greater on the painful side and whether this
translation causes symptoms (49). We have found
laxity assessment by drawer testing to be of limited value in the
office setting because patients tend to guard the painful shoulder.
Instead, this method is used during examination under anesthesia to
confirm the suspected degree and direction of shoulder instability.

Altchek et al. (3) and Hawkins et al. (65)
proposed a grading scale for translation of the humeral head on the
glenoid. Instability is graded on a scale of 0–3+ for all three
directions. For anterior and posterior drawer testing, a grade of 0
represents no humeral head translation, while movement of the humeral
head up to but not over the glenoid rim represents 1+ instability.
Translation of the humeral head over the glenoid rim with an associated
spontaneous reduction with relief of pressure represents

2+ instability. Frank dislocation and locking of the humeral head over the glenoid rim is graded as 3+ instability.

When these drawer tests are performed either in the
office or in the operative setting, it is important to bear in mind
that the position of the arm determines the degree of tension in the
glenohumeral ligaments. With the arm at the side in adduction, the IGHL
is relatively lax, and anterior and posterior drawer testing may be of
limited value. In abduction, the IGHL comes underneath the humeral head
and forms a hammock that passively limits anterior, posterior, and
inferior translation (113,114).
Perform anterior drawer testing with the shoulder positioned in
abduction in the plane of the scapula. Maintain the arm in the neutral
rotation while using one hand to place an axial load along the humerus
and the other hand to apply an anterior or posterior force to the
humerus (Fig. 80.11). Often you can feel the
humeral head move back into the glenoid rather than out of the glenoid
during this maneuver. The patient may note a painful click with such a
maneuver. In our experience, this test is particularly helpful in
identifying posterior instability.

A modification of the posterior drawer test allows the
examiner to elicit posterior apprehension. Perform this modification by
placing the patient’s arm in 90° of forward flexion and adduction while
applying an axial load down the shaft of the humerus. Pain and a
palpable shift and click suggest posterior labral injury and
instability.

A modification of this test, termed the jerk test, has been described for posterior instability (12,86).
With the patient seated, load the abducted shoulder axially into the
glenoid with one hand, and with your other hand, palpate the posterior
aspect of the shoulder. Then bring the arm into horizontal adduction
anterior to the plane of the scapula; the humeral head may sublux
posteriorly. Then bring the humerus posterior to the plane of the
scapula; the humeral head may suddenly reduce into the glenoid. A
palpable shift and pain accompany a positive test.

The sulcus sign is basically an inferior drawer test (Fig. 80.12). Neer and Foster (107)
originally described it as the hallmark of inferior and
multidirectional instability (MDI). Unfortunately, a common
misconception has been that a large sulcus sign that is asymptomatic,
thus indicating inherent joint laxity, is a positive finding. The key
point is that this maneuver should be associated with pain and should
reproduce the patient’s symptoms to be clinically relevant as a finding
of inferior instability.

Perform this test with the patient seated and the arm
adducted at the side. Rotation of the shoulder is very important in
assessing the degree of inferior instability. First, with the arm in
the neutral rotation, pull the humerus inferiorly, and estimate the
amount of separation between the acromion and the humeral head. Grade
it on a 0–3+ scale (3): A separation of 1 cm is
a 1+ sulcus sign, 2 cm is a 2+ sulcus sign, and 3 cm is a 3+ sulcus
sign. Anatomically, a sulcus sign greater than 2+ indicates a capacious
capsule and specific laxity of the anterosuperior capsular region
(rotator interval).

As previously discussed, externally rotate the arm, and
repeat the sulcus sign test. If the sulcus sign remains greater than 2+
with the arm in external rotation, there is a marked deficiency of the
superior capsule, and a large rotator interval defect in the capsule is
likely. This is the result of damage to the superior and middle
glenohumeral ligaments, as well as the coracohumeral ligament. With
this information before surgical repair, you then know that surgical
reconstruction of this region with a capsular shift must be a component
of the operation (60).

Four different types of MDI patterns have been described (119):

Type I: global instability

Type II: predominantly anterior and inferior instability

Type III: predominantly posterior and inferior instability

Type IV: anterior and posterior instability with no inferior component

The true existence of the last type has been questioned.

Since the description of superior labral pathology by Andrews et al. (5) in 1985 and of the SLAP lesion by Snyder et al. (145)
in 1990, several examination techniques have evolved to diagnose this
pathology. Andrews reported increased pain in patients during full
shoulder flexion and abduction, with noticeable catching and popping.
Snyder reported pain in patients during resisted shoulder flexion with
elbow extension and forearm supination (biceps tension test).

Another useful diagnostic test is the
compression–rotation test. With the patient supine, abduct the shoulder
90°, with the elbow flexed 90°. Apply a compression force to the
humerus to trap the torn labrum (in the same manner as McMurray’s test
for the knee is performed). O’ Brien et al. (114)
suggested a maneuver that they believed is accurate for determining
superior labral injuries. Place the patient’s shoulder in 90° of
forward flexion and then adduct it across the body. Ask the patient to
flex the arm further against resistance when the shoulder is first
internally rotated and then externally rotated. If pain occurs when the
shoulder is rotated internally but not when it is rotated externally,
the test is positive. Unfortunately, this test does not distinguish
between acromioclavicular joint disease and a superior labral tear. We
have found this test more useful than the other methods described above.

The minimum radiography necessary for evaluation of an
acute dislocation or suspected subluxation is a true anteroposterior
(AP) view and an axillary view. These images will allow accurate
determination of the position of the humeral head relative to the
glenoid. A true AP radiograph is obtained by angling the x-ray beam 45°
relative to the sagittal plane of the body (Fig. 80.13). A scapular Y or transcapular view can also give useful information about the position of the humeral head (Fig. 80.14), but it is not as accurate as an axillary view (Fig. 80.15).
If a standard axillary view cannot be obtained, a Velpeau axillary view
without removing the patient’s arm from the sling will suffice (Fig. 80.16).

In the office setting, a true AP view of the shoulder
with the arm in internal rotation may demonstrate a Hill–Sachs lesion.
A Stryker notch view is a special view that will also demonstrate a
Hill–Sachs lesion (Fig. 80.17) (43,121).

The West Point axillary view may prove helpful in a
patient suspected of having had an episode of instability. Take the
image with the patient prone so that the anterior glenoid is shown in
profile without an overlying acromial shadow (Fig. 80.18).

Adjuvant imaging techniques are helpful when additional information about the three-dimensional relationship

and architecture of the joint or confirmation of the presence of a
Bankart lesion is required. Computed tomography (CT) demonstrates bony
injuries or abnormalities including glenoid hypoplasia, congenital
version anomalies, acquired version abnormalities from erosion, and
glenoid rim fractures (Fig. 80.1 and Fig. 80.19). In addition, it allows measurement of the size of a humeral head defect (Hill–Sachs lesion) in cases of chronic instability (55). When combined with intraarticular dye, CT arthrography also demonstrates a Bankart lesion and articular erosion.

Magnetic resonance imaging (MRI) with or without
gadolinium has gained great favor, although unfortunately it is often
used as a screening tool in the evaluation of patients. Its role should
instead be to confirm the presence of lesions that may need surgical
management. It is accurate in the detection of labral pathology (Fig. 80.20) (27,56).
Recent advances in MRI technology such as high-resolution fast-spin and
gradient-echo imaging make these methods particularly appealing for
detecting anterior and superior labral injuries (26).

Before arthroscopy or arthrotomy is undertaken,
examination under anesthesia (EUA) can confirm or deny the preoperative
impression (30,154).
During EUA, muscle guarding is eliminated, and an accurate assessment
of joint laxity is possible. As previously described, it is essential
that laxity not be confused with instability; thus, it is important to
compare the symptomatic shoulder with the asymptomatic side.
Individuals who have had a previous dislocation will likely have a 2+
or 3+ anterior drawer test result (as described above) for the
symptomatic shoulder compared with a 1+ – 2+ finding in the
asymptomatic shoulder.

The finding of symmetric laxity, however, does not
necessarily exclude instability because a Bankart lesion may be present
without marked laxity on examination in a patient who has a history of
subluxation (149,171).
Furthermore, in very large individuals, the soft-tissue envelope of the
shoulder may be so large that accurate determination of a centimeter of
translation of the humeral head on the glenoid may not be possible.

One area where we have found EUA to be helpful is in determining capsular laxity and rotator interval lesions (45),
both of which are typically associated with MDI. If an individual is
found to have a sulcus sign of more than 2+ and if this persists when
an inferior drawer is applied with the arm in external rotation, there
will be a large rotator interval defect, as well as marked anterior and
inferior capsular laxity. In our hands, this is treated by a modified
capsular shift, along with repair of a Bankart lesion if one is found
to be present (60). Although other surgeons
have used arthroscopic techniques in the treatment of this combination
of problems, we generally treat only pure Bankart lesions with an
arthroscopic repair technique (99,136,159).

Occasionally, EUA will identify a shoulder that can be dislocated posteriorly (3+)
when anterior instability is suspected. It is usually in a patient with
congenital laxity who has had a traumatic injury. We usually treat this
condition with an open capsular shift method of repair.

Portal placement is extremely important when an arthroscopic Bankart repair is performed.

As seen in the algorithm in Figure 80.32,
there are many errant courses that result in ineffective treatment. A
patient or a physician can make these errors. A successful outcome
begins with an accurate diagnosis followed by the specific treatment
required for that particular patient.

Detrisac and Johnson (39)
originally described five variations in labral anatomy; more recently,
however, Detrisac (personal communication) changed the categories to
two basic types. Type A labra were attached centrally and peripherally
at the inferior, posterior, and anterior regions, but superiorly they
were detached centrally and peripherally. In contrast, type B labra
were attached centrally and peripherally along the entire labrum.

Snyder et al. (146), who
performed shoulder dissections, modified this description. They
described an inferior labrum that consistently was triangular in shape
and attached centrally and peripherally. A similar shape was
demonstrated in the posterior labrum, although occasionally the
undersurface of this area was not attached. The superior region once
again showed consistent attachment peripherally, with a central portion
that was free in varying degrees. The anterior labrum was attached at
both the central and peripheral regions, appearing as a thickened band
of capsule that was clearly distinguishable on histologic and gross
assessment.

Several patterns of labral injury have been described.
Flap tears are usually found in the anterosuperior, anteroinferior, or
inferior labrum (146), associated with
subluxation or frank dislocation. Other labral tears that have been
described include bucket-handle and incomplete split tears, which have
not necessarily been associated with instability, if located superiorly
(120). Degenerative tears are labral lesions that coexist with degenerative joint disease.

Although initial reports on arthroscopic labral debridement were encouraging (4,54,122),
later reports demonstrated a clear relationship between instability and
labral pathology. In each study, despite early favorable results at 1-
and 2-year follow-up, there was significant deterioration in clinical
outcome. The authors in each article concluded that athletes whose
sports involve overhead activity should be treated as if there is some
underlying laxity or instability. Mere debridement of the labral
pathology cannot be expected to be successful.

A study of 52 patients undergoing consecutive arthroscopic labral debridement at Columbia-Presbyterian Medical

Center further strengthened these recommendations (35).
The majority of these patients had SLAP lesions, while the remainder
had either anteroinferior or posterior labral lesions. While no patient
had a history of dislocation or clinically evident instability
preoperatively, EUA revealed instability in 70% of patients with SLAP
lesions and in all of the other patients.

At 1-year follow-up, 78% of the superior labral group
versus 30% of the anteroinferior group had excellent results from
debridement. These results deteriorated with time to 63% and 25% in
each respective group at 2-year follow-up. The authors concluded that
instability was frequently present in patients with labral pathology,
and debridement would provide only short-term relief in this patient
population. On the basis of these studies it is difficult to recommend
labral debridement for most patients.

In 1985, Andrews et al. (5)
reported the association in athletes between throwing and injuries to
the anterosuperior glenoid labrum and the insertion of the long head of
the biceps brachii. These injuries were believed attributable to the
large deceleration forces applied by the biceps muscle at the insertion
point. Two studies further implicated the long head of the biceps in
contributing to anterior and superior shoulder stability during
abduction and external rotation (126,174). In 1990, Snyder et al. (145)
retrospectively reviewed 700 patients who underwent shoulder
arthroscopy and described four specific types of pathologic
abnormalities in this area: the SLAP lesions (Fig. 80.33). A description of each type of SLAP lesion and the recommended treatment is provided in Table 80.5.

Anatomic and histologic studies further characterized the superior labral area (21,31,129).
These studies demonstrated a relative avascularity at the superior and
anterosuperior regions compared with the posterior and inferior
regions, and this may be one reason for the degeneration that occurs in
this area. A retrospective review by Snyder et al. (144)
involving 2,375 arthroscopic shoulder procedures over 8 years found
SLAP lesions in 140 (6%) patients. Most lesions were classified as
either type II (55%) or type I (21%) lesions. Only 28% had isolated
superior

labral
pathology, 22% had associated Bankart lesions, and the remainder had
associated rotator cuff pathology. Another review of 712 patients
revealed that 17% had type I lesions, while 11.8% had other significant
labral pathology (91).
The authors had difficulty classifying the remaining pathology
according to the criteria described by Snyder, but once again there was
a high association between labral lesions and subtle instability or
laxity on EUA.

The correlation between glenohumeral instability and
superior-labral and biceps-insertion pathology has led to the
development of several arthroscopic repair techniques. In 1991, Yoneda
et al. (184) described the use of an
arthroscopic staple (Instrument Makar Inc., Okemos, MI) designed by
Lanny Johnson. The patient population consisted of 10 young athletes
(involved in overhead sports) with type II SLAP lesions. The standard
lateral decubitus position was used, and the joint was visualized
through the standard posterior portal. An anterosuperior portal was
used for instrumentation.

The surgery involved arthroscopic debridement, abrasion,
and stapling of the SLAP area. One staple (5.5 mm diameter) was used in
each case. Patients were immobilized for 2–3 weeks; range of motion was
then initiated, followed by strengthening exercises at 6 weeks. Because
of concerns over staple migration and irritation, all staples were
removed at an average of 3.9 months. On second-look arthroscopy,
complete healing had occurred in four patients, with full continuity
with the glenoid surface. The remaining repairs were deemed stable
despite incomplete healing. More important, 80% of patients were found
to have good-to-excellent results.

Reports of loosening and irritation with the arthroscopic staple have led to the development of alternative repair techniques (36,60,124).
A transglenoid approach, originally developed for Bankart lesions and
popularized by Caspari, has been applied to the SLAP lesion (24,44).

Some concerns have been raised about this technique, however (166),
because it requires a separate posterior incision to tie the sutures.
Tying over the infraspinatus muscle may lead to gaps at the repair site
if muscle atrophy occurs. Further, the elastic nature of PDS suture
makes it

prone to gap 5–10 mm at the repair site (52). In addition, during transosseous pin placement, there is a risk of injury to the suprascapular nerve or one of its branches (140).

Many of the arthroscopic techniques for SLAP lesion
repairs were adapted from techniques for arthroscopic repair of Bankart
lesions. Arthroscopic shoulder stabilization was initially attempted
with an arthroscopic staple (79).
Unfortunately, this technique was associated with a high recurrence
rate and an unacceptably high rate of complications related to the
implant (Table 80.6). Complications included
breakage and migration of the staple, as well as articular and
neurovascular injuries from the staple. This technique has largely been
abandoned.

Subsequently, Caspari and Morgan developed a
transosseous suture repair technique that avoided the pitfalls of a
juxtaarticular staple (24a,104,136). This method was based on the open technique developed by Reider and Ingles (166).
In a manner similar to that previously described for SLAP lesion
repairs, a Caspari suture punch is used to place PDS sutures through
the IGHL. These sutures are then loaded onto a special pin with a hole
at one end (Beath pin) and passed through a drill hole made at the
anterior glenoid rim so that it exits out of the infraspinatus fossa (Fig. 80.34). These sutures are then separated into two limbs and tied over the infraspinatus fascia.

In the hands of the surgeons who developed this method
of repair, results have been good to excellent over intermediate
follow-up (157). Unfortunately, others have
been unable to duplicate the success of these surgeons, and with longer
follow-up, recurrence rates have beenreported in the 20% to 40% range (Table 80.7) (58,127,140,162,165).

As discussed previously, there are technical problems
associated with this technique, such as placement of the drill hole too
medially so that the Bankart lesion is not repaired, as well as
suprascapular nerve injury from posterior exit of the pin. Finally, the
need for an additional posterior incision has made this approach less
desirable.

One of the first attempts to develop a reliable all intraarticular

technique was made by Maki (92).
He proposed drilling across the glenoid in a manner similar to the
technique described above; however, a large suture (Mulberry knot) was
tied in the PDS suture so that when it was pulled back it locked on the
posterior cortex of the scapula. Internal knots were then tied over the
anterior capsule and labrum with a special knot pusher and special
knot-tying techniques. Current approaches to arthroscopic Bankart
repair involve the use of suture anchors and internal knots or
absorbable transfixing implants (42,127,183).

Regardless of which implant is used, arthroscopic
methods of repair are continuing to evolve, along with an appreciation
of more specific indications for their use. There are clear-cut
advantages and disadvantages to arthroscopic stabilization procedures (Table 80.8).
Appropriate patient selection is based on a detailed history and
physical examination, imaging studies when indicated, and a careful EUA
and arthroscopic inspection. In addition, other specific factors seem
to be associated with a good result (Table 80.4).

While some surgeons are exploring the role of
arthroscopic capsulorrhaphy and thermal capsular shrinkage for patients
with capsular laxity, we feel that the optimal patient for repair is
one who has a discrete Bankart lesion with robust capsuloligamentous
structures without significant capsular injury. In addition, it is
preferable if there is no significant glenoid erosion or a bony Bankart
lesion. While a small Hill–Sachs lesion is not a contraindication to
arthroscopic repair, we believe that a large Hill–Sachs lesion is
associated with recurrent episodes of dislocation and generally
indicates significant capsular laxity. With these criteria, short-term
follow-up studies have demonstrated a significant reduction to
approximately 7% to 10% in postoperative failure rates

(recurrences) after arthroscopic stabilization procedures (84,175,176).

Since Neer and Foster’s (107) original description of MDI, many open techniques have been developed to stabilize patients with the condition (3,10,78,171).
While results with these methods have been good, they have not been on
par with the results after treatment of traumatic unidirectional
anterior instability; return to overhead athletic activities has been
less certain. Moreover, there is a perception that perioperative
morbidity might be improved with a “less invasive” arthroscopic
technique. For this reason, several surgeons have attempted to develop
arthroscopic methods of capsulorrhaphy to treat patients with MDI.
While some surgeons have reported on their initial experience with
these methods (Eugene Wolf and Steven Snyder, personal communications),
little has been reported in the peer-reviewed literature.

Duncan and Savoie (41) described an arthroscopic inferior capsular shift in 1993. The technique involved dividing

the anterior and inferior capsule with an end-cutting shaver. This
division allowed the inferior capsule to be advanced superiorly.
Sutures were tied with a transglenoid suture technique. This initial
report was a preliminary study of 10 patients who were all found to
have a “satisfactory” shoulder according to the Neer and Foster
criteria. An average Rowe score of 90 was reported, as well.
Modifications of this transglenoid suture capsulorrhaphy have been
developed to address posterior instability (100).

Wichman and Snyder (180)
described a technique of capsular plication in a select population with
MDI and no associated labral pathology. All patients underwent a
prolonged course of rehabilitation before surgical intervention, which
allowed serial examinations. Over the weeks, the physicians’ knowledge
of the patient and the underlying disorder increased. Thus, patients
with voluntary instability of a psychological nature could be
identified.

With this technique, external rotation was preserved to
within 10° of the preoperative level in 92% of patients at a minimum
2-year follow-up. An assessment of postoperative function using the
American Shoulder and Elbow Society instrument revealed an improvement
from a preoperative score of 63 to a postoperative score of 86.2 The
average Rowe score postoperatively was 79.6. Five patients’s conditions
were rated unsatisfactory, with 19 patients having satisfactory
ratings, according to Neer’s criteria. The authors noted a
disproportionately high number of workers’ compensation cases in the
unsatisfactory group. Interestingly, three patients underwent
second-look arthroscopy (two for unrelated procedures) that revealed
significant capsule formation at the site of previous plication and no
adverse synovial reaction to the suture.

The use of heat to treat shoulder instability dates back
to 460 BC when Hippocrates inserted a hot iron into the axilla,
contracting the inferior aspect of the joint. This same concept has
gained recent popularity, but laser and radiofrequency energy sources
are used to produce the heat. The triple-helix structure of collagen
unwinds at 65°C, undergoing a conformational change from a crystalline,
extended state to a randomly coiled, contracted state (denatured) and
producing shrinkage in capsular tissues (106). Current literature has focused on laser technology as a source of the thermal energy. Hayashi et al. (66) demonstrated a dose-dependent effect of nonablative laser energy on capsular tissue properties.

The Holmium:yttrium-aluminum-garnet (Ho:YAG) laser has received considerable attention (1,141).
Laser energy at low power—less than 10 watts (1.0 joule per pulse at 10
hertz)—is recommended for laser-assisted capsular shrinkage. There are
several disadvantages to the use of laser energy. First, lasers are
very expensive and require a certain level of expertise and appropriate
certification for use. Further, lasers produce significant power that
can be difficult to control.

Radiofrequency thermal stimulation has been used as an
alternative energy source that avoids some of the problems inherent in
laser energy. Such devices are less expensive and are relatively easier
to use. The ElectroThermal Arthroscopy system (ORATEC Interventions,
Inc., Menlo Park, CA) integrates a thermocouple into the arthroscopic
handpiece, creating a monopolar system. The current applied to the
tissue at the dull probe tip is minimally invasive while still
accomplishing collagen denaturation (S.P. Arnoczky, personal
communication). This system contrasts with the vapor system from Mitek
Surgical Products, Inc. (Westwood, MA), which represents a bipolar
system without a thermocouple. Both products are radiofrequency systems.

These instruments have been applied over the past
several years as an experimental treatment method. Their obvious appeal
is that they can be used effectively in an outpatient setting. The
procedure does not require a great deal of time or the use of sutures
or implants. Unfortunately, while anecdotal reports concerning these
procedures are appearing, there is currently no peer-reviewed
literature available regarding the effectiveness of this technique.
Nevertheless, thermal capsular shrinkage remains a possible alternative
treatment. Its efficacy will need to be substantiated over the next
5–10 years with well controlled follow-up studies to assess whether
there is a gradual restretching of the tissues with increased use.

The indications for open versus arthroscopic shoulder stabilization remain a controversial topic (164). As outlined in Table 80.4,
the criteria for arthroscopic shoulder stabilization are becoming more
clearly defined. As surgeons continue to improve their ability to
arthroscopically shift the glenohumeral ligaments and repair labral
lesions and as new technologies are developed, the number of patients
who are candidates for open surgical procedures will decrease.
Certainly, the presence of a large Hill–Sachs lesion or a bony Bankart
lesion on radiographic examination is an indication for an open
surgical approach. Likewise, patients without a significant Bankart
lesion by MRI or CT may not be optimal candidates for an arthroscopic
Bankart procedure. Similarly, findings consistent with significant
capsular laxity or stretch, such as an atraumatic history or MDI
findings on physical examination, should direct the surgeon toward open
treatment options. Significant capsular damage, rotator interval
lesions, or rotator cuff tears are also currently better repaired with
open surgical techniques.