Between 1988 and 1991, 241 shoulder decompressions were performed
by the senior author (J.R.A.). Thirty-nine (16%) of these
operations were consecutive resections of the distal aspect of the
clavicle with concomitant subacromial decompression. Seven patients
(seven shoulders) were lost to follow-up, leaving thirty-one patients
(thirty-two shoulders) to form the basis for this study. During the
same period, twelve isolated resections of the distal aspect of
the clavicle were performed.
There were twenty-nine men and two women in the study. The mean
age of the patients at the time of surgery was thirty-six years
(range, eighteen to sixty-seven years), and the mean age at the
time of follow-up was forty-one years (range, twenty-two to seventy-one
years). The dominant shoulder was involved in twenty-seven patients.
The mean duration of follow-up was four years and ten months (range,
three to eight years). Twenty-five of the patients, including four
professional athletes, were actively involved in sports activities.
Preoperative diagnoses included shoulder impingement and symptomatic
acromioclavicular pathology in all of the patients. Acromioclavicular
lesions included atraumatic osteolysis of the distal aspect of the
clavicle in three shoulders (two shoulders of weight-lifters and
one shoulder of a pitcher), traumatic osteolysis of the distal aspect
of the clavicle in one shoulder, type-I acromioclavicular joint injury
in ten shoulders, type-II acromioclavicular joint injury in five
shoulders, and acromioclavicular joint arthritis in thirteen shoulders.
Intraoperative findings included a partial rotator cuff tear
in ten shoulders, one large rotator cuff tear, and two so-called
SLAP lesions (superior labrum, anterior and posterior) (types I
and III). Additional intraoperative findings included proliferative
subacromial bursitis in twenty-nine shoulders, thickening and fibrosis
of the subacromial bursa in thirteen shoulders, an acromial protuberance
in thirty-two shoulders, rotator cuff tendinitis with inflammation of
the cuff in twelve shoulders, posterior labral fraying in seven
shoulders, and capsular synovitis in nine shoulders.
Indications for distal clavicular resection included continued
pain that was recalcitrant to conservative therapy for a minimum
of six months, a positive crossover adduction test, tenderness to
palpation of the acromioclavicular joint, and a good response to a
lidocaine injection. Criteria for a concomitant subacromial decompression
were a positive impingement sign, a positive lidocaine impingement
test, and evidence of subacromial impingement on arthroscopic and
radiographic examination. The mean amount of time from the onset
of symptoms to the resection was ten months (range, six to twenty-six
months).
Acromioclavicular joint degeneration, distal clavicular osteolysis,
and acromioclavicular separation (types I and II) as seen radiographically
were considered relative indications for surgery when associated
with acromioclavicular pain36,37.
A type-II separation was considered a relative indication if it
was determined that the acromioclavicular pain was caused by incongruity
and/or degeneration of the joint rather than by instability. In
addition, a type-II or type-III acromion with periarticular calcifications
was considered a relative indication for subacromial decompression
when associated with clinical signs of impingement.
The follow-up evaluation included clinical examination, chart
review, radiographic analysis, and isokinetic strength-testing of
both upper extremities. All of the patients were followed for a
minimum of two years. Clinical results were evaluated with a previously
reported assessment method38.
A two-tailed paired t test was used to compare preoperative and
postoperative shoulder functional scores.
Clinical evaluation was performed by examiners (S.D.M. and T.E.B.)
who were not involved in the patient’s surgery. Subjective
evaluation of the patient’s satisfaction with the results
and the ability to return to athletics involving overhead activity was
carried out. Specifically, patients were asked whether they were
satisfied with the result, whether shoulder function on overhead
activity had returned to its previous level, and whether they would
have the procedure again.
On preoperative physical examination, all of the patients had
a positive crossover adduction test, tenderness on direct palpation
of the acromioclavicular joint, and a positive shoulder impingement sign.
A diagnostic lidocaine test for acromioclavicular joint disease
was performed in twenty-eight shoulders, and it was positive in
all of them. In addition, a lidocaine impingement test was performed,
during an office visit separate from the visit at which the acromioclavicular
joint injection was given, on twenty-five shoulders, and it was positive
in each of them. Six patients (seven shoulders) had had a previous
positive lidocaine injection test performed by the referring physician.
Shoulder-joint strength-testing was performed with a Biodex isokinetic
dynamometer (Biodex, Shirley, New York) at velocities of 60° and
180°/sec for flexion and extension testing. Rotational
isokinetic strength-testing was carried out at 180° and 300°/sec.
Patients had undergone at least two prior tests on the dynamometer
to familiarize them with the machine and the testing sequence. Shoulder
flexion and extension as well as external and internal rotation
in the scapular plane were tested.
Routine preoperative and postoperative radiographs included an
anteroposterior view of the shoulder with a 30° caudal tilt, a Y
scapular view with a 10° to 15° caudal tilt (outlet view), an axillary
view, and a Zanca view of the acromioclavicular joint (an anteroposterior
view of the acromioclavicular joint with a 10° cephalic tilt and
50% penetrance)39. In
addition, a previously described profile view was made of all of
the patients40. This radiograph
is made by placing the patient supine on the x-ray table (Figs. 1-A, 1-B, and 1-C). The involved
extremity is brought across the upper part of the abdomen with the
elbow flexed. Some patients may need to place the hand on the contralateral
shoulder to further elevate the involved shoulder; however, this
is usually not necessary for muscular patients. The contralateral extremity
reaches across the forehead to grasp a 20 ¥ 25-cm cassette.
The cassette lies pressing into the midpoint of the trapezius and
is parallel to the sagittal plane of the body. The central ray is
directed toward the humeral head at an angle 20° away from the perpendicular
line. It enters 4 cm below the humeral head at a distance of 100
cm. In extremely obese patients, the angle is increased 2° to 5°. Immediately
before exposure, the patient is asked to depress the cassette into
the soft belly of the trapezius.
An anterior acromial protuberance, defined as a portion of the
acromion projecting anterior to the anterior border of the distal
aspect of the clavicle identified on the axillary radiograph, was
present in twenty-two shoulders. In addition, sclerotic bone reaction
of the greater tuberosity was noted in nineteen shoulders; sclerosis
of the anterior aspect of the acromion, in four shoulders; osteophytes
in the inferior portion of the acromioclavicular joint, in thirteen
shoulders; degenerative changes of the acromioclavicular joint,
in twenty-two shoulders; osteolysis of the distal aspect of the
clavicle, in four shoulders; and calcification of the supraspinatus tendon,
in two shoulders.
Surgical Technique
After the induction of general anesthesia, physical examination
of both shoulders was performed to document range of motion, shoulder
laxity, and acromioclavicular joint stability. The patient was placed
in the lateral decubitus position with the torso supported by a
vacuum beanbag. The arm was suspended at 70° of abduction and 15°
of forward flexion and was held in place with a prefabricated wrist
gauntlet or soft wrap. The position of the arm was maintained with
an overhead pulley system and a counterweight of fifteen or twenty
pounds (6.8 or 9.1 kg). Once the patient was securely positioned, the
shoulder and arm were aseptically prepared and draped. The wrist
gauntlet was covered with a sterile towel and a plastic drape.
Diagnostic arthroscopy was begun through a posterior portal.
An arthroscopic fluid pump was used in all cases. Arthroscopic examination
of the glenohumeral joint was conducted in a systematic fashion
to identify and inspect all of the intra-articular structures regardless
of the preoperative diagnosis. Partial cuff tears (ten shoulders)
were treated with local débridement. Two patients (two
shoulders) had a SLAP lesion (types I and III), which was also treated
with débridement.
After the glenohumeral joint was evaluated and treated, subacromial
bursoscopy was carried out through the same portals. A 4.5-mm cannula
with a blunt trocar was redirected through the posterior skin incision
toward the posterolateral edge of the acromion. The undersurface
of the posterolateral aspect of the acromion was palpated with the
trocar, and the cannula with the trocar was advanced into the anterolateral
aspect of the subacromial space to avoid bleeding.
A lateral instrument portal was established approximately 3 cm
from the lateral edge of the acromion in approximately the midcoronal
plane of the acromion. An 18-gauge needle was inserted into the subacromial
space, and, once it was visualized, a small skin incision was made
at the point of insertion. Instruments were inserted directly through
the lateral skin portal (Fig. 2-A). Electrocautery was used to maintain
hemostasis and to remove soft tissue from the undersurface of the
acromion. The decompression was carried out with a 5.5-mm acromionizer
burr maintaining an anterior sleeve of periosteum, with resection
of the anterior aspect of the acromion to the coronal level of the
anterior edge of the distal aspect of the clavicle. After acromioplasty,
the coracoacromial ligament was transected with electrocautery.
Upon completion of the subacromial decompression, the 30° scope
remained in the posterior portal and electrocautery was introduced
through the lateral portal to remove soft tissue and fat from around the
undersurface of the acromioclavicular joint, including the inferior
aspect of the acromioclavicular joint capsule. Electrocautery was
also used to strip periosteum from the distal aspect of the clavicle,
exposing the area to be resected.
The acromioclavicular joint was identified with an 18-gauge spinal
needle passed from above the joint. The soft-tissue shaver was used
to remove all fibrous tissue from the medial border of the acromion
and acromioclavicular joint region. A 5.5-mm burr was then introduced
through the lateral portal to begin resection of the distal aspect
of the clavicle (Fig. 2-B). Next, 8 to 10 mm of bone was
resected from the distal aspect of the clavicle. The undersurface
of the distal aspect of the clavicle was burred level with the subacromial
decompression. At this point, the burr was introduced into the anterior
portal and the 30° scope was switched to a 70° scope for improved
upward visualization of the distal aspect of the clavicle during
resection. Exposure was further enhanced by manually depressing
the distal aspect of the clavicle during resection and switching
the arthroscope to the lateral portal for a more direct view of
the acromioclavicular joint region2.
Steri-Strips (3M, St. Paul, Minnesota) were applied to the arthroscopic
portals, and the involved extremity was placed in a sling. The sling
was removed one day postoperatively, and physical therapy was initiated.
The patients were seen for clinical follow-up at one week, six weeks,
and three months postoperatively or until full function returned.
The mean functional score (and standard deviation) for individual
activities was 2.7 0.5 points (range, 2.1 to 3.0 points) preoperatively
and 3.9 0.2 points (range, 3.6 to 4.0 points) postoperatively (p = 0.0001).
Multiple linear-regression analysis was performed to identify potential
predictors of total functional scores by considering the age of
the candidates in the model as well as the gender, diagnosis, acromioclavicular
joint stability, presence of a rotator cuff tear, soft-tissue calcification,
amount of bone resected, and type of acromion. A stepwise multiple
linear-regression procedure did not show any of these variables
to be predictive of the mean functional score (p > 0.20
in each case). In addition, the mean functional scores were compared
between type-I acromioclavicular joint injuries (ten shoulders)
and type-II (five shoulders), and no differences were found with
a two-sample t test. However, the sizes of these subgroups were
too small for us to make any definitive statements.
All of the patients were satisfied with their result and had
improvement in their functional result. Of the twenty-five patients
who participated in sports, twenty-two (including the four professional
athletes) returned to their previous level of sports activity.
Twenty-six patients had no more pain, three patients had mild
pain on strenuous repetitive overhead activity, and two patients
(both weight-lifters) noted occasional pain on strenuous overhead
activity (bench-pressing). In addition, two patients (both baseball
players) had mild pain in the posterior aspect of the shoulder on
throwing.
Two of the four dedicated weight-lifters in the study had slightly
less strength in the involved arm; this affected the maximum amount
that they could bench-press, but the difference was not detectable on
clinical evaluation or isokinetic testing. No other patient noted
weakness after the procedure. The patient with a large cuff tear
had a grade of 4 (of 5) on manual strength-testing of external rotation
and abduction. No other patient had detectable weakness postoperatively
on manual strength-testing.
No patient lost motion as a result of the procedure. Five patients
had some limitation of passive internal rotation compared with that
of the contralateral shoulder. This limitation averaged 10° when
measured with the arm at 90° of abduction, and four of these patients
were able to reach a point two thoracic levels more caudad and the
fifth, one level more caudad, when passive internal rotation was measured
with the arm at the side. This was unchanged from the motion on
the preoperative examination. No patient changed occupations as
a result of the shoulder surgery.
On postoperative radiographic analysis, no patient had superior
migration of the clavicle when compared with its preoperative position.
The mean amount of distal clavicular resection was 9 mm (range,
7 to 15 mm) (Figs. 3-A, 3-B, and 3-C). One patient had heterotopic ossification
at the resection site, with mild pain on direct palpation of the
acromioclavicular joint. This was one of the patients who had noted
mild pain on strenuous overhead activity. In addition, five patients
had calcification at the anterior deltoid insertion into the acromion.
All of these patients were asymptomatic, with no impingement on
overhead activity and no pain on direct palpation. One patient had
mild impingement with extremes of forward flexion and abduction
that was exacerbated by internal rotation, as described by Hawkins
and Kennedy41. No other patient
had a positive impingement sign postoperatively.
Isokinetic testing revealed a mean peak torque-to-body weight
ratio in flexion of 4% more than that of the uninjured
shoulder at 60°/sec and 8.5% more at 180°/sec.
The mean peak torque-to-body weight ratio in extension was 9% more
at 60°/sec and 9% more at 180°/sec than
that of the uninjured shoulder.
On rotational isokinetic strength-testing at 180°/sec the
mean peak torque-to-body weight ratio in external rotation was 8% stronger
than that of the uninjured shoulder, and at 300º/sec
it was 13% stronger. On testing of internal rotation at
180°/sec the peak torque-to-body weight ratio was 2% stronger
than that of the uninjured shoulder, and at 300°/sec it was
1% weaker.
Neer emphasized the importance of proliferative spurs on the
undersurface of the anterior aspect of the acromion in his article
on impingement syndrome in 197242.
Neer and Poppen described the supraspinatus outlet—consisting
of the space between the acromion, coracoacromial ligament, coracoid,
acromioclavicular joint, and glenoid—through which the supraspinatus
muscle passes33. They considered
the primary cause of shoulder impingement to be narrowing of this
outlet. Neer popularized the use of an open two-stage anterior acromioplasty
to ensure removal of the offending anterior osteophyte in addition
to any anterior-inferior proliferation of the acromion or the distal aspect
of the clavicle42,43.
In 1985, Ellman described arthroscopic anterior acromioplasty4. Since then, various authors have
demonstrated favorable results of arthroscopic subacromial decompression1,5,6,10,13-15. However, successful
treatment of shoulder impingement is directly proportional to the
accuracy of the diagnosis44. In
addition to insufficient acromioplasty and rotator cuff disease,
failure of acromioplasty can sometimes be attributed to persistent
acromioclavicular symptoms16,45;
this led Neviaser et al. to recommend routine excision of the acromioclavicular
joint at the time of acromioplasty46.
Chronic causes of acromioclavicular pain include an idiopathic,
intra-articular disc pathology, posttraumatic degenerative arthrosis
from joint incongruity, primary degenerative arthrosis, and rheumatoid
arthrosis. Osteolysis of the distal aspect of the clavicle is associated
with repetitive microtrauma of the acromioclavicular joint from activities
such as weight-lifting, gymnastics, and swimming20,26,27.
The underlying pathophysiology is believed to be an inflammatory
process with hyperemic resorption of the distal aspect of the clavicle20,27. Other causes of distal clavicular
osteolysis include rheumatoid arthrosis, hyperparathyroidism, and sarcoidosis,
which should be considered in the differential diagnosis, especially
in bilateral cases47,48.
Open resection of the distal aspect of the clavicle as a treatment
option for chronic acromioclavicular pain was initially reported
independently by both Gurd49 and
Mumford50, with good results.However, on occasion, substantial
morbidity, such as disruption of the deltotrapezial fascia and anterior
aspect deltoid rupture, can occur29,31,48,51.
Arthroscopic resection of the distal aspect of the clavicle has
been described as having results similar to those of open resection3,4,7,10,23,25,28,30,52.
Advantages include a smaller surgical scar with preservation of the
acromioclavicular ligaments, capsule, and deltotrapezial fascial
attachments to the clavicle. This encourages accelerated rehabilitation
with immediate motion, a shorter hospital stay, and possibly a quicker
return to functional and athletic activities3,7,30,52.
Arthroscopic evaluation of the glenohumeral joint and rotator cuff
may be performed to rule out concomitant pathology23,25.
There have been relatively few studies documenting subacromial
decompression with acromioclavicular joint resection, with many
of the reported cases included in larger series of isolated subacromial decompression16,45,53. Thorling et al. reported
that the results in eleven patients with distal clavicular resection
at the time of acromioplasty were inferior to the results in forty patients
who underwent acromioplasty alone35.
Jalovaara et al. reported satisfactory results of acromioplasty
in 88% of cases and satisfactory results of acromioplasty
combined with distal clavicular resection in 76%11. Flatow et al. noted no adverse
effects of acromioplasty combined with distal clavicular resection
and reported satisfactory results in 83% of fifty-five cases8. However, they noted a 50% failure
rate, with continued instability in shoulders with grade-II acromioclavicular
separation8.
We do not recommend routine acromioclavicular joint resection
at the time of subacromial decompression. In most patients with
shoulder impingement syndrome undergoing decompression, only the
osseous excrescences on the undersurface of the joint need to be
removed. However, in cases where a symptomatic acromioclavicular
joint pathology is contributing to shoulder pain that is recalcitrant
to conservative therapy, arthroscopic resection with concomitant
subacromial decompression can provide excellent functional results
with good pain relief. As with other shoulder disorders, an accurate clinical
diagnosis, proper patient selection, and a sufficient trial of conservative
treatment are essential for a successful outcome.