Abstract
Background: Neer and Foster previously described the inferior
capsular shift procedure for treating multidirectional instability
of the shoulder and reported preliminary results that were quite satisfactory.
The purpose of our study was to perform a longer-term follow-up
evaluation of the efficacy of the inferior capsular shift procedure
for treating multidirectional instability of the shoulder.
Methods: An inferior capsular shift procedure
was used to treat multidirectional instability of the shoulder in forty-nine
patients (fifty-two shoulders). All patients had failed to respond
to an exercise program. In this series, the operative approach (anterior
or posterior) was based on the major direction of the instability,
as determined by the preoperative history and physical examination
and as verified by examination with the patient under anesthesia.
In all of the patients, the inferior capsular shift was the primary
attempt at operative stabilization. The repair consisted of a lateral-side
(or humeral-side) shift of the capsule to reduce capsular redundancy and,
when necessary, a reattachment of the avulsed labrum to the anteroinferior
aspect of the glenoid.
Results: A redundant capsular pouch was seen
in all of the shoulders in this series. In addition, detachment
of the anteroinferior aspect of the labrum was found in ten shoulders
and an anterior fracture of the glenoid rim was seen in two shoulders.
At an average of sixty-one months (range, twenty-four to 132 months),
results were available for forty-nine shoulders (forty-six patients).
Thirty shoulders (61 percent) had an excellent overall result, sixteen
(33 percent) had a good result, one (2 percent) had a fair result,
and two (4 percent) had a poor result. Forty-seven (96 percent)
of the forty-nine shoulders remained stable at the time of follow-up.
Two of the thirty-four shoulders that had been repaired through
an anterior approach began to subluxate anteroinferiorly again.
None of the fifteen shoulders that had been repaired through a posterior
approach had recurrent instability. Full function, including the
ability to perform strenuous manual tasks, was restored to forty-five
shoulders (92 percent). A return to sports was possible after thirty-one
(86 percent) of the thirty-six procedures done in athletes; however,
a return to the premorbid level of participation was possible after
only twenty-five (69 percent) of the thirty-six procedures.
Conclusions: The results in this series demonstrate
the efficacy and the durability of the results of the inferior capsular
shift procedure for the treatment of shoulders with multidirectional
instability. The procedure directly addresses the major pathological
feature - a redundant joint capsule. Similar results were seen with
either an anterior or a posterior approach, and we continue to approach
shoulders with multidirectional instability on the side of greatest
instability. A postoperative brace was reserved for patients in
whom a posterior approach had been used or in whom an anterior approach had
involved extensive posterior capsular dissection (ten of the thirty-four
shoulders treated with the anterior approach).
Multidirectional instability of the shoulder is instability in
all three directions: anterior, posterior, and inferior. Neer and
Foster pointed out that it is important to distinguish this entity
from unidirectional instability because the standard repairs designed
to correct unidirectional instability will fail when performed on
a multidirectionally unstable shoulder16,17.
Specifically, a unidirectional anterior or posterior repair will
not correct the major pathological feature in multidirectional instability
- namely, excessive redundancy of the capsular ligaments, especially
inferiorly. Thus, inferior subluxation may persist after a unidirectional
repair16-18,20,21,26. Moreover,
overtightening one side of a multidirectionally unstable shoulder
can exacerbate the instability in the direction left unaddressed.
This can result in recurrent instability in the opposite direction
or in fixed subluxation, leading to early degenerative arthritis8,9,13,16,22,24.
In 1980, Neer and Foster described the inferior capsular shift
procedure for treating patients with symptomatic multidirectional
instability of the shoulder who had failed to respond to nonoperative
treatment16. Satisfactory results
were achieved with this repair in thirty-nine of forty shoulders,
though only short-term follow-up data was available in that preliminary
report. Cooper and Brems also reported satisfactory results, at
a mean of thirty-nine months, after treating multidirectional instability with
an inferior capsular shift procedure5.
In their series, the procedure was performed exclusively through
an anterior surgical approach. Others have described arthroscopic
techniques for performing the inferior capsular shift procedure6,14.
The purpose of our study was to perform a longer-term follow-up
evaluation of the efficacy of the inferior capsular shift procedure
in the management of multidirectional instability of the shoulder.
Between 1982 and 1992, we performed an inferior capsular shift
procedure in fifty-two shoulders in forty-nine patients with multidirectional
instability. Three of these patients were lost to follow-up less
than two years after the procedure, leaving forty-nine shoulders
in forty-six patients for consideration in this study. The average
duration of follow-up was sixty-one months (range, twenty-four to
132 months). Since many of the patients with multidirectional instability
seen over this ten-year period responded satisfactorily to a program of
rotator cuff and scapular muscle strengthening exercises, the patients
in this series represent a selected group who had not responded
to extensive efforts at nonoperative treatment before undergoing
surgical repair. In order to be considered for this study, a shoulder
had to exhibit classic multidirectional instability (that is, instability
in the anterior, posterior, and inferior directions) as demonstrated
by the history, an office physical examination, and an examination
with the patient under anesthesia at the time of repair. Specifically, the
anterior apprehension and posterior stress tests19 as
well as the sulcus sign16 were
helpful in categorizing the instability during the office examination.
On examination with the patient under anesthesia, these shoulders
could be translated over the glenoid rim anteriorly and posteriorly
and had a large sulcus with inferior stress. Patients with voluntary
or willful instability were excluded from the study. The inferior
capsular shift procedure was the initial attempt at operative stabilization
in all of these shoulders.
There were twenty-eight male and eighteen female patients in
the study. The average age at the time of repair was twenty-three
years, with a range of sixteen to forty-two years. Twenty-seven
right shoulders and twenty-two left shoulders were involved. Three
patients underwent a bilateral (staged) inferior capsular shift
procedure. The dominant arm was involved in twenty-nine patients
(63 percent).
A major episode of trauma was involved with the onset of symptoms,
or the first instability event, in twenty-five shoulders (51 percent).
Recurrent microtrauma, such as that occurring with overhead sports
activities, was believed to be the major etiological factor in nine
other shoulders (18 percent), though many of the patients whose
shoulder became unstable after trauma had also participated extensively
in overhead sports activities. In fifteen shoulders (31 percent),
the onset of instability was entirely unassociated with trauma.
Twenty-four (49 percent) of the shoulders dislocated anteriorly and
inferiorly with subluxation posteriorly. Fifteen (31 percent) of
the shoulders dislocated posteriorly and inferiorly with subluxation
anteriorly. Ten shoulders (20 percent) dislocated in all three directions.
Thirty-four shoulders (69 percent) underwent the inferior capsular
shift procedure through an anterior approach as the anteroinferior
component of the instability was predominant or the shoulder dislocated
in all three directions. In this group, the symptoms predominantly
occurred with the arm in combined abduction and external rotation.
The anterior apprehension test, performed by progressively externally
rotating and extending the shoulder with the arm abducted to 90
degrees and the elbow flexed to 90 degrees, was positive in all of
these patients. A positive sulcus sign, signified by the appearance
of a dimple or sulcus inferior to the lateral aspect of the acromion
with downward traction on the neutrally positioned arm, was seen in
thirty shoulders (88 percent) in the office and in all thirty-four
shoulders on examination with the patient under anesthesia at the
time of repair. In addition, ten of these shoulders could be easily
dislocated posteriorly when the shoulder was flexed to 90 degrees,
adducted, and internally rotated, and this posterior stress test
reproduced instability symptoms. On examination with the patient
under anesthesia, these ten shoulders tended to stay posteriorly
subluxated over the glenoid rim with a posteriorly directed force
on the shoulder at 0 degrees of flexion. The remaining twenty-four
shoulders that were treated through the anterior approach could
be subluxated posteriorly, but they then spontaneously reduced when
the posterior manual force was removed during the examination with
the patient under anesthesia.
Fifteen shoulders (31 percent) underwent the inferior capsular
shift procedure through a posterior approach as the posteroinferior
component of the instability was predominant. In this group, symptoms
occurred most often with the arm in combined flexion, adduction,
and internal rotation, as occurs during the follow-through phase
of throwing or swimming or with activities of daily living such
as removing a pullover shirt. The posterior stress test, performed
by the examiner stabilizing the scapula with one hand and applying
a posteriorly directed force to the humerus, which is flexed to
90 degrees, adducted, and internally rotated, produced subluxation
or dislocation with symptoms in fourteen of the fifteen shoulders. Eleven
of the fifteen shoulders had a positive sulcus sign on examination
in the office, and all fifteen had a positive sign on examination
with the patient under anesthesia. Seven shoulders had a positive
anterior apprehension test preoperatively. In one shoulder, the
examination with the patient under anesthesia demonstrated marked
posterior instability whereas the preoperative office examination
had failed to do so. This resulted in a change in the operative
approach from anterior to posterior. However, this was the only
shoulder in the series for which the planned operative approach
was changed at the time of surgery on the basis of the examination
with the patient under anesthesia.
Method of Evaluation
The shoulders were evaluated on the basis of stability, pain,
range of motion, and activity level3.
The result for all shoulders that had a recurrence of instability
after repair was designated as a failure. Stability was assessed
on the basis of the history and with provocative testing, though provocative
testing was deferred for nine months postoperatively. Stable repairs
were then evaluated with use of an 8-point rating system to assess pain,
range of motion, and activity level. An overall rating of 8 points
indicated an excellent result; one of 6 or 7 points, a good result;
one of 4 or 5 points, a fair result; and one of less than 4 points, a
poor result or a failure. No pain was graded as 4 points; mild pain
after strenuous use, as 3 points; mild-to-moderate pain after active
use, as 2 points; and severe pain, as 0 points. Active motion was graded
as full (2 points), satisfactory (1 point), or poor (0 points).
Full active motion was defined as elevation in the scapular plane
of at least 160 degrees, external rotation of at least 60 degrees,
and internal rotation (measured as the most cephalad level that
the thumb could reach on the spine) to the tenth thoracic level
or cephalad. Satisfactory active motion was defined as elevation
of 130 to less than 160 degrees, external rotation of 30 to less
than 60 degrees, and internal rotation to a level between the fifth
lumbar and tenth thoracic vertebrae but not reaching the tenth thoracic
vertebra. Poor active motion was defined as elevation of less than
130 degrees, external rotation of less than 30 degrees, and internal
rotation to a level caudad to the fifth lumbar vertebra. As for
the activity level, a full return to activities, including strenuous
manual tasks and sports activities (when applicable), was given
2 points; a return to normal activities of daily living but some
limitation with regard to strenuous manual tasks, 1 point; and limitations with
activities of daily living, 0 points.
Operative Technique
All patients in the series underwent an inferior capsular shift
procedure, as originally described by Neer and Foster2,16. As discussed above, the operative
approach (anterior or posterior) was based on the predominant direction
of instability in each case, which in turn was determined by the
preoperative symptoms and physical findings and was verified at
the time of surgery with an examination with the patient under anesthesia.
Anterior Approach
With the patient supine in a modified beach-chair position, a
concealed anterior axillary skin incision, measuring approximately
six to eight centimeters, is made in an axillary skinfold12. The incision starts several centimeters
inferior to the coracoid tip and proceeds in an axillary skinfold
to the inferior border of the pectoralis major muscle. The deltopectoral
interval is developed, with the cephalic vein retracted laterally
with the deltoid. The clavipectoral fascia is incised lateral to
the combined origin of the coracobrachialis and the short head of
the biceps, and these muscles are gently retracted medially. The
superior and inferior borders of the subscapularis muscle are identified,
and the anterior circumflex humeral vessels at its inferior border
are coagulated. The subscapularis tendon is incised one centimeter
medial to its insertion on the lesser tuberosity, with the incision
proceeding from the rotator interval superiorly to the inferior
border of the tendon. The plane between the subscapularis tendon
and the underlying capsular ligaments is developed medially to the
glenoid rim, and the tendon is tagged with nonabsorbable sutures
for later repair. Care is taken to avoid perforating the capsule
during this portion of the dissection. The dissection proceeds inferiorly, and
the muscular position of the subscapularis is also separated from
the capsule, with the axillary nerve identified and protected. Complete
separation of the subscapularis muscle and tendon from the capsule
is essential, lest the capsule become tethered and the ability to
shift the capsule be limited.
The capsule is then incised 0.5 centimeter medial to the subscapularis
cut, starting superiorly in the region of the capsular cleft between
the superior and middle glenohumeral ligaments and proceeding inferiorly
(Fig. 1-A).
A sufficient cuff of tissue is left laterally on the neck of the
humerus, to which the capsular flaps will subsequently be repaired.
The dissection proceeds inferolaterally along the humeral neck to avoid
cutting the inferior pouch in half, which would render the flaps
unsuitable for a capsular shift and would jeopardize the axillary
nerve. The dissection proceeds inferiorly, until the redundant inferior
pouch can be sufficiently reduced by pulling up on the traction
sutures placed in the capsule and thus extruding the surgeon's index
finger from the redundant inferior pouch (Fig. 1-B). In shoulders
with multidirectional instability, this usually requires dissecting
down past the six o'clock position on the humeral neck. When the capsule
has been sufficiently mobilized, a ring retractor is used to retract
the humeral head to allow inspection of the anteroinferior aspect
of the labrum. When the glenohumeral ligaments have been avulsed
from their labral attachment, they are reattached to the anteroinferior
aspect of the glenoid rim, before the surgeon proceeds with the capsular
shift. In order for the shift to be effective, the capsule must
be anchored medially to the glenoid. In this series, this was accomplished
by using nonabsorbable number-0 braided nylon sutures passed through
osseous tunnels; more recently, suture anchors have also been employed
for this purpose. Next, the capsule is split in T-fashion almost to
its labral insertion, just above the superior border of the inferior
glenohumeral ligament. This creates two capsular flaps: a superior
flap consisting of the middle glenohumeral ligament and an inferior
flap consisting of the inferior glenohumeral ligament. The arm is
then placed in approximately 20 degrees of abduction and 25 to 30 degrees
of external rotation. The inferior flap is pulled superiorly, reducing
the inferior capsular pouch, and is sutured to the lateral capsular
remnant (Fig. 1-C).
The capsular cleft between the superior and middle glenohumeral
ligaments is closed, and this entire superior flap is then shifted
inferiorly over the inferior flap in cruciate fashion to reinforce
the capsule anteriorly (Fig. 1-D). The subscapularis is then repaired
to its anatomical insertion. The deltopectoral interval is closed, and
finally the skin is closed with a subcuticular suture.
Postoperative care: In the present series, twenty-four
(71 percent) of the thirty-four repairs that were performed through an
anterior approach were followed by protection of the arm in a sling.
In this group of twenty-four shoulders, range-of-motion exercises
were begun after ten to fourteen days and were gradually progressed.
External rotation was limited to less than 30 degrees for the first
six weeks after repair. The patients were seen frequently in the
early postoperative period, and the range of motion was carefully
monitored; those who were regaining motion too quickly were considered
to be at risk for stretching out the repair, and their protocol
was adjusted accordingly. Isometric exercises were started three
weeks after surgery, but generalized strengthening exercises for
the deltoid, rotator cuff, and scapular stabilizers were deferred
for the first six weeks postoperatively. The remaining ten shoulders
that were operated on through an anterior approach, in which there
had been posterior dislocations prior to surgery and that were found
to have an especially redundant posteroinferior aspect of the capsule
at surgery, were treated with a brace, with the arm positioned in
neutral rotation at the side, for six weeks postoperatively. The postoperative
rehabilitation protocols described above were then instituted and
adapted for each individual shoulder. Return to sports always was
delayed for a minimum of six months and usually was delayed for
nine to twelve months, especially if the patient participated in
high-demand overhead sports.
Posterior Approach
This approach is performed with the patient in the lateral decubitus
position, with the involved shoulder pointing upward. An oblique
skin incision directed 60 degrees from the scapular spine is used. The
incision starts at the posterolateral corner of the acromion and
extends distally for approximately ten centimeters. The deltoid
is split along a posterolateral raphe for a distance of four to
five centimeters. The deltoid is also detached from the scapular
spine for a distance of three to four centimeters in order to facilitate
exposure (Fig. 2-A).
A cuff of deltoid is left on the scapular spine for later repair
of the deltoid. The infraspinatus is identified and differentiated
from the supraspinatus superiorly and from the teres minor inferiorly. The
infraspinatus is then separated from the underlying posterior aspect
of the capsule, medially past the rim of the glenoid and laterally
to its insertion on the greater tuberosity. The infraspinatus tendon may
be incised obliquely, starting medially in a superficial plane and
progressing laterally and deeply, thus creating two tendon flaps,
as originally described16. However,
if the infraspinatus tendon is too thin to allow this, it is incised
vertically, leaving a stump of tissue laterally for later repair
of the tendon.
The capsule is then incised one centimeter medial to its insertion
on the humerus, starting superiorly and proceeding inferiorly. The
plane between the capsule and the teres minor is developed bluntly, and
the axillary nerve is protected. The capsule is dissected around
the humeral neck as far inferiorly as necessary to allow reduction
of the posteroinferior capsular redundancy. As with the anterior approach
described above, this dissection usually proceeds well past the
six o'clock position on the humeral neck in shoulders with multidirectional instability.
The arm is extended and internally rotated to allow better exposure
of this region of the inferior aspect of the capsule. Again, as
with the anterior approach, the surgeon places an index finger into
the pouch to determine if enough capsule has been mobilized. If
the finger can be extruded by pulling up on the traction sutures
in the capsule, then sufficient mobilization has been accomplished
to allow an appropriate decrease in joint volume by shifting of
the capsule. At this point, if the posterior aspect of the labrum
is found to be detached, which is unusual, it is repaired back to the
glenoid rim, prior to shifting of the capsule. The capsule is then
split in T-fashion at the mid-glenoid region, creating a superior
and an inferior flap. The superior capsular flap is shifted inferiorly and
is reattached with nonabsorbable braided nylon sutures to the cuff
of tissue left on the lateral aspect of the humeral neck (Fig. 2-B). The inferior
flap is then shifted superiorly to reduce the inferior pouch and
to reinforce the posterior aspect of the capsule (Fig. 2-C). The arm
is positioned in 5 to 10 degrees of external rotation, 10 to 15
degrees of abduction, and neutral flexion-extension during tensioning
of the capsular flaps. The infraspinatus is then repaired with number-1
or 2 nonabsorbable braided nylon sutures. The deltoid is reattached
to the scapular spine, and the posterolateral split in the deltoid
is repaired. The skin incision is closed with a subcuticular suture.
Postoperative care: All of the patients who underwent the inferior
capsular shift procedure through a posterior approach wore a plastic
brace postoperatively for approximately six weeks. The brace consisted
of a plastic shell with an outrigger to support the shoulder in
slight abduction and neutral rotation, thus preventing stresses
on the posterior and inferior aspects of the capsule. Range-of-motion
exercises, emphasizing elevation in the scapular plane and external
rotation, were usually begun six weeks postoperatively. Terminal
flexion and internal rotation were avoided for approximately three
months. Isometric exercises were also begun six weeks after surgery,
and there was gradual progression to a fuller program of strengthening
exercises. Sports activities were restricted for nine to twelve
months after surgery.
Operative Findings
A redundant capsular pouch was seen in all forty-nine shoulders
in this series. In addition, detachment of the anteroinferior aspect
of the labrum was found in ten shoulders. A fracture of the anterior aspect
of the glenoid rim was found in two shoulders. In both of these
shoulders, the fragment was small, measuring less than 10 to 15
percent of the articular surface. All twelve of the anteroinferior labral
or anterior osseous lesions occurred in shoulders that had demonstrated
predominantly anteroinferior instability and had been treated through
an anterior surgical approach. In only one shoulder, in which a
posterior approach was employed, was a posterior labral detachment
found.
Operative Results
Thirty shoulders (61 percent) had an excellent overall result,
sixteen (33 percent) had a good result, one (2 percent) had a fair
result, and two (4 percent) had a poor result. Thus, forty-six shoulders
(94 percent) had an overall successful (good or excellent) result.
Forty-seven (96 percent) of the forty-nine shoulders remained
stable at the time of the most recent follow-up. Two shoulders that
had undergone repair through an anterior approach began to subluxate
anteroinferiorly again, and these shoulders were rated as failures.
None of the shoulders that had been repaired through a posterior
approach had recurrent instability.
Thirty-three shoulders (67 percent) were essentially pain-free
at the time of follow-up, whereas fifteen (31 percent) were mildly
painful after strenuous use, such as with sports activities. Only one
shoulder (2 percent) remained moderately painful after repair. Forty-seven
shoulders (96 percent) had a full active range of motion postoperatively,
and the remaining two (4 percent) had a satisfactory range of motion
but persistent limitation in external rotation. None of the shoulders were
categorized as having poor active motion.
Forty-five shoulders (92 percent) regained a full level of function,
including that required for strenuous manual tasks. Only four shoulders
(8 percent) had persistent functional limitations with regard to strenuous
use, but they could be used for all simple activities of daily living.
Thirty-one (86 percent) of thirty-six shoulders in athletes recovered enough
function for the patient to be able to return to participation in
the premorbid sports activities. These activities included contact
sports, such as football and wrestling, as well as high-demand overhead
sports, such as baseball, swimming, and tennis. However, only twenty-five
(69 percent) of the thirty-six shoulders functioned well enough
for the patient to return to the premorbid level.
Anterior Approach
Of the thirty-four shoulders that underwent repair through an
anterior approach, twenty-two (65 percent) had an excellent result,
nine (26 percent) had a good result, one (3 percent) had a fair
result, and two (6 percent) were rated as failures. Thus, thirty-one
(91 percent) of the thirty-four shoulders had an overall satisfactory
rating.
Thirty-two (94 percent) of the thirty-four shoulders remained
stable at the time of final follow-up. Two shoulders had a recurrence
of the anteroinferior instability: one of them, which had had recurrent
dislocations prior to repair, began to subluxate, and the other,
which had had recurrent subluxations prior to repair, began to subluxate again.
Twenty-three shoulders (68 percent) were essentially pain-free
at the time of follow-up. Ten (29 percent) were mildly painful after
strenuous use, and one (3 percent) caused persistent moderate pain,
which resulted in a fair rating. Thirty-three (97 percent) of the
shoulders had a full range of motion at the time of follow-up, and
the remaining shoulder (3 percent) had satisfactory motion with residual
restriction in external rotation.
Thirty-one shoulders (91 percent) had full, unrestricted function,
whereas the other three (9 percent) had limitations with regard
to strenuous use but functioned well enough for all routine activities
of daily living. Twenty-three (85 percent) of twenty-seven shoulders
in athletes functioned well enough for the patient to return to
participation in his or her sport, and twenty-one (78 percent) of
the twenty-seven functioned well enough for the patient to return
to the premorbid level.
Posterior Approach
Of the fifteen shoulders treated through a posterior approach,
eight had an excellent result and seven had a good result. There
were no failures in this group. All fifteen of these shoulders were
stable at the time of follow-up.
Ten shoulders were pain-free at the time of follow-up, whereas
five caused mild pain after strenuous use. None of these shoulders
had moderate or severe pain postoperatively.
Fourteen shoulders had a full range of motion postoperatively,
and one shoulder had satisfactory motion. Fourteen of the fifteen
shoulders regained full, unlimited function, and the remaining shoulder
had limitations with regard to strenuous use but could be used to
perform all routine activities of daily living. Eight of the nine
shoulders in athletes functioned well enough for the patient to
return to participation in sports, but only four functioned well
enough for the patient to return to the premorbid level. Throwers
and swimmers specifically noted a decrease in their level of performance
compared with their premorbid ability.
The results of the present study support earlier findings5,11,15,16 concerning the efficacy
of the inferior capsular shift procedure for treating shoulders
with multidirectional instability that had failed to respond to
a program of rehabilitative exercises4.
This procedure directly addresses the pathological capsular redundancy
found in these shoulders. Unlike some of the older repairs for the
treatment of instability (for example, the Putti-Platt and Magnuson-Stack
procedures), which effected an asymmetrical unidirectional tightening
of the shoulder and allowed persistent inferior subluxation or subluxated
the humeral head in the opposite direction, the inferior capsular
shift allows for a symmetrical reduction of capsular redundancy and
joint volume16,17. The lateral
capsular approach employed in this repair is suggested by the funnel-shaped
anatomy of the joint capsule, with a much broader insertion on the
humeral side than on the glenoid side. Thus, more capsule can be
shifted, and it can be shifted farther on the lateral (or humeral)
side. This is particularly important in cases of multidirectional
instability, in which the capsular redundancy may be global (involving
the anterior, posterior, and inferior aspects of the joint).
This study demonstrated that the results of the inferior capsular
shift procedure are durable, a finding that supports the data previously
reported by Cooper and Brems5.
In their preliminary report on the procedure, Neer and Foster reported
success in thirty-nine of forty shoulders, but only seventeen had
been followed for more than two years16.
Indeed, while the inferior capsular shift procedure has been widely
recommended for the treatment of multidirectional instability, there
has been relatively little information about the longer-term results
of this repair. In the present study, after an average duration
of follow-up of more than five years, only two shoulders (4 percent)
became unstable again, and in both cases this occurred within the
first postoperative year. Even with resumption of high-demand sports
activities by many of these patients, there were no late failures
in this series.
A review of the demographic data in this series is instructive
with regard to several points. The stereotype of patients with multidirectional
instability as thin, sedentary females with wholly atraumatic instability
was not borne out by the patients in this study. Sixty-one percent
of the patients were male, and 72 percent were involved in athletics
at the time of the onset of the symptoms. Moreover, a major episode
of trauma was involved with the onset of symptoms, or the first
instability event, in half of the shoulders in the study. More than
one etiological factor appeared to underlie the expression of the
instability in many of these patients; these factors included repetitive
microtrauma superimposed on an inherently loose shoulder and trauma
superimposed on one or both of those etiologies. Thus, any classification
of instability must take into account that multiple etiological factors
may be at play in a given shoulder. Such an understanding will help
to avoid the underdiagnosis of multidirectional instability in,
for example, athletic patients, many of whom do not fit the stereotype
of the patient with multidirectional shoulder instability.
Review of this data also helps to clarify the role of the examination
with the patient under anesthesia at the time of surgery. This examination
verified the surgeon's preoperative impression about the predominant
direction of the instability and thus about the surgical approach
(anterior or posterior) in forty-eight (98 percent) of the forty-nine
shoulders. In only one instance did the examination contradict the
preoperative findings and lead the surgeon to change the intended
operative approach. However, the examination performed with the
patient under anesthesia helped to quantify more precisely the degree
of laxity, particularly in the inferior direction. Muscle-guarding
during the performance of the sulcus test in the preoperative office
examination not infrequently minimized the magnitude of the inferior
laxity; the examination with the patient under anesthesia allowed
a more accurate measurement. In a number of shoulders, only a small
sulcus (less than one centimeter) could be elicited during the office
examination, whereas with the patient under anesthesia a large sulcus
(more than two centimeters) could be demonstrated in all of these
shoulders. Thus, the examination with the patient under anesthesia facilitated
more complete measurement of the laxity in each direction but rarely
changed the clinical impressions obtained from a carefully recorded preoperative
history and a thorough physical examination.
Another interesting finding was the presence of an avulsion of
the anteroinferior aspect of the labrum, a so-called Bankart lesion,
in ten shoulders (20 percent) and an anterior fracture of the glenoid
rim in two others (4 percent). In the original description of the
inferior capsular shift procedure16,
the authors purposely excluded from the study patients with a Bankart
lesion who had been operated on through an anterior approach and
included only those in whom the Bankart lesion had been discovered
during repair of capsular redundancy through a posterior approach.
In the present study, we chose to include all patients with signs
and symptoms of multidirectional instability, regardless of whether
the lesion was isolated to the capsule or included labral avulsion.
In the shoulders with both labral avulsion and capsular redundancy,
the labrum was first reanchored to the glenoid rim to restore the
medial integrity of the capsulolabral mechanism. In addition, however,
a symmetrical reduction of capsular redundancy was performed in
these multidirectionally unstable shoulders. None of the anteroinferior
labral avulsions occurred in shoulders that had been operated on
through a posterior approach, and thus a second incision was not
required in any patient.
Some authors have chosen to perform capsular reconstruction for
multidirectional instability exclusively through an anterior operative
approach5,25. In our series, we
approached the shoulder on the side of greatest instability as demonstrated
by the preoperative history and physical examination and as confirmed
at the time of surgery by the examination with the patient under
anesthesia. In this manner, the region of maximum capsular redundancy
is most directly approached anteroinferiorly with the anterior approach
and posteroinferiorly with the posterior approach. We believe that,
when the pathology is primarily posteroinferior capsular redundancy,
the posterior approach allows better visualization and access to
this region of the capsule. Similar results were achieved with the
anterior and posterior approaches, and no recurrent instability
was seen in shoulders that had been repaired through the posterior
approach. Similar levels of success with posterior capsulorrhaphy have
been reported recently in other series of patients with posterior
instability (which included, but were not limited to, those with
multidirectional instability)3,7,10,19,23.
The only difference between the anterior and posterior-approach
groups was that the patients treated with the posterior approach
had a diminished ability to return to overhead sports activities
at the premorbid level. We continue to employ the posterior approach
in patients in whom the instability is predominantly in the posteroinferior
direction, but we emphasize that, regardless of the approach, an
essential feature of this procedure is adequate separation of the
muscle from the capsule inferiorly to prevent tethering and to allow
effective shifting of the capsule and reduction of joint volume.
Although previous authors have recommended that all patients
be treated with a spica cast or a brace for six weeks after an inferior
capsular shift procedure15,16,
we reserved the use of a brace for two groups: patients who had
undergone the inferior capsular shift through a posterior approach
and those with previous posterior dislocations who had been treated
through an anterior approach. For these patients, positioning the
arm in neutral rotation (as opposed to internal rotation in a standard
sling) avoids tension on the posterior aspect of the repaired capsule.
Shoulders with multidirectional instability in which the major components
were anterior and inferior and that had only a lesser degree of
posterior subluxation were protected in a standard sling for six
weeks. These patients required less dissection of the posteroinferior
aspect of the capsule, and thus it was not deemed necessary to avoid
internal rotation. Limited range-of-motion exercises were generally
begun in this group after ten to fourteen days. For all of the patients,
the rate of progression of the rehabilitation program was carefully
monitored by the surgeon. Full motion was restored in nearly all
of these patients, but care was taken to do this gradually, in order
to avoid stretching out the repair. For a similar reason, athletic
activities, particularly those involving throwing or swimming, were
delayed for nine to twelve months postoperatively.
In conclusion, the inferior capsular shift procedure is effective
for treating multidirectional instability of the shoulder, with
a high percentage of satisfactory results at the time of long-term
follow-up. The operative approach, either anterior or posterior, was
chosen on the basis of the direction of greatest instability, and
similar results were achieved with both approaches in this series.
A major technical consideration in this procedure, regardless of whether
it is performed through an anterior or posterior approach, is careful
separation of the capsule from the overlying musculotendinous layer
(the subscapularis anteriorly or the infraspinatus and teres minor
posteriorly) in order to allow effective shifting of the ligaments
to reduce capsular redundancy. The type of postoperative immobilization (brace
or sling) and the arm position during this period were chosen according
to the degree of posterior capsular dissection. Finally, postoperative rehabilitation
was individualized for each shoulder but generally proceeded more
gradually than that designed for patients who have undergone repair because
of more unidirectional instability.
Altchek, D. W.; Warren, R. F.; Skyhar, M. J.; and Ortiz, G.: T-plasty modification of the Bankart procedure for multidirectional
instability of the anterior and inferior types. J. Bone and Joint Surg.,73-A: 105-112, Jan 1991.73-A105
1991
Bigliani, L. U.: Anterior and posterior capsular shift for multidirectional
instability. Tech. Orthop.,3: 36-45, 1989.336
1989
Bigliani, L. U.; Pollock, R. G.; McIlveen, S. J.; Endrizzi, D. P.; and Flatow, E. L.: Shift of the posteroinferior aspect of the capsule for
recurrent posterior glenohumeral instability. J. Bone and Joint Surg.,77-A: 1011-1020, July 1995.77-A1011
1995
Burkhead, W. Z., Jr., and Rockwood, C. A., Jr.: Treatment of instability of the shoulder with an exercise
program. J. Bone and Joint Surg.,74-A: 890-896, July 1992.74-A890
1992
Cooper, R. A., and Brems, J. J.: The inferior capsular-shift procedure for multidirectional
instability of the shoulder. J. Bone and Joint Surg.,74-A: 1516-1521, Dec 1992.74-A1516
1992
Duncan, R., and Savoie, F. H., III: Arthroscopic inferior capsular shift for multidirectional
instability of the shoulder: a preliminary report. Arthroscopy,9: 24-27, 1993.924
1993
[PubMed]
Fronek, J.; Warren, R. F.; and Bowen, M.: Posterior subluxation of the glenohumeral joint. J. Bone and Joint Surg.,71-A: 205-216, Feb 1989.71-A205
1989
Hawkins, R. H., and Hawkins, R. J.: Failed anterior reconstruction for shoulder instability. J. Bone and Joint Surg.,67-B(5): 709-714, 1985.67-B(5)709
1985
Hawkins, R. J., and Angelo, R. L.: Glenohumeral osteoarthrosis. A late complication of the
Putti-Platt repair. J. Bone and Joint Surg.,72-A: 1193-1197, Sept 1990.72-A1193
1990
Hawkins, R. J., and Janda, D. H.: Posterior instability of the glenohumeral joint. A technique
of repair. Am. J. Sports Med.,24: 275-278, 1996.24275
1996
[PubMed]
Lebar, R. D., and Alexander, A. H.: Multidirectional shoulder instability. Clinical results
of inferior capsular shift in an active-duty population. Am. J. Sports Med.,20: 193-198, 1992.20193
1992
[PubMed]
Leslie, J. T., Jr., and Ryan, T. J.: The anterior axillary incision to approach the shoulder
joint. J. Bone and Joint Surg.,44-A: 1193-1196, Sept 1962.44-A1193
1962
Lusardi, D. A.; Wirth, M. A.; Wurtz, D.; and Rockwood, C. A., Jr.: Loss of external rotation following anterior capsulorraphy
of the shoulder. J. Bone and Joint Surg.,75-A: 1185-1192, Aug 1993.75-A1185
1993
McIntyre, L. F.; Caspari, R. B.; and Savoie, F. H., III: The arthroscopic treatment of multidirectional shoulder
instability: two-year results of a multiple suture technique. Arthroscopy,13: 418-425, 1997.13418
1997
[PubMed]
Mizuno, K.; Itakura, Y.; and Muratsu, H.: Inferior capsular shift for inferior and multidirectional
instability of the shoulder in young children: report of two cases. J. Shoulder and Elbow Surg.,1: 200-206, 1992.1200
1992
Neer, C. S., II, and Foster, C. R.: Inferior capsular shift for involuntary inferior and multidirectional
instability of the shoulder. A preliminary report. J. Bone and Joint Surg.,62-A: 897-908, Sept 1980.62-A897
1980
Neer, C. S., II: Involuntary inferior
and multidirectional instability of the shoulder: etiology, recognition,
and treatment. In Instructional Course Lectures, American Academy
of Orthopaedic Surgeons. Vol. 34, pp. 232-238. Park Ridge, Illinois, American
Academy of Orthopaedic Surgeons, 1985.
Norris, T. R., and Bigliani, L. U.:
Analysis of failed repair for shoulder instability - a preliminary
report. In Surgery of the Shoulder, pp. 111-116. Edited by J. E.
Bateman and R. P. Welsh. Philadelphia, B. C. Decker, 1984.
Pollock, R. G., and Bigliani, L. U.: Recurrent posterior shoulder instability. Diagnosis and
treatment. Clin. Orthop.,291: 85-95, 1993.29185
1993
[PubMed]
Rockwood, C. A., Jr., and Gerber, C.: Analysis of failed surgical procedures for anterior shoulder
instability. Orthop. Trans.,9: 48, 1985.948
1985
Rowe, C. R.; Zarins, B.; and Ciullo, J. V.: Recurrent anterior dislocation of the shoulder after surgical
repair. Apparent causes of failure and treatment. J. Bone and Joint Surg.,66-A: 159-168, Feb 1984.66-A159
1984
Samilson, R. L., and Prieto, V.: Dislocation arthropathy of the shoulder. J. Bone and Joint Surg.,65-A: 456-460, April 1983.65-A456
1983
Santini, A. M., and Neviaser, R. J.: Analysis of long term results of posterior inferior capsular
shift [abstract]. J. Shoulder and Elbow Surg.,5: 512, 1996.5512
1996
Steinmann, S. R.; Flatow, E. L.; Glasgow, M. D.; and Bigliani, L. U.: Evaluation and surgical treatment of failed shoulder instability
repairs. Orthop. Trans.,16: 727, 1992-1993.16727
1992-1993
Wirth, M. A.; Groh, G. I.; and Rockwood, C. A., Jr.: Capsulorraphy through an anterior approach for the treatment
of atraumatic posterior glenohumeral instability with multidirectional
laxity of the shoulder. J. Bone and Joint Surg.,80-A: 1570-1578, Nov 1998.80-A1570
1998
Young, D. C., and Rockwood, C. A., Jr.: Complications of a failed Bristow procedure and their
management. J. Bone and Joint Surg.,73-A: 969-981, Aug 1991.73-A969
1991