Arthroscopic subacromial decompression (ASAD) is becoming a widely performed surgical procedure of the shoulder. The technique has evolved from open anterior acromioplasty as described by Neer,1-2 Hawkins et al,3’4 Rockwood,5 and Bigliani et al.6 The transition from open to arthroscopic technique entails a definite learning curve and should not be underestimated. This chapter focuses on the technical aspects of the procedure and how to avoid complications.
The arthroscopic technique for subacromial decompression was first described by Johnson7 in 1986. Ellman8 presented the first series with follow-up and detailed description of the operative technique. Esch et al9 evaluated their results with ASAD and related them to the severity of associated rotator cuff tears. Paulos and Franklin10 presented one of the largest early series (80 patients) and introduced the use of the midlateral subacromial portal.
All of these authors originally described the procedure with the scope viewing from the posterior portal and the instruments entering from a lateral approach.
Sampson et al11 first described the “cutting block” technique for precision acromioplasty in 1991. This technique places the scope laterally and introduces shaving and burring instruments from a posterior portal, using the posterior half of the acromion as a guide for resection. The authors also emphasized the importance of the supraspinatous outlet x-ray in both preoperative planning and postoperative evaluation and the benefits of evaluating the flatness of the cut from both the lateral and the posterior portals.
Many orthopaedists (myself included) who began performing arthroscopic acromioplasty from the originally described lateral approach now routinely utilize a technique incorporating the cutting-block principles. There are, however, still a number of cases where the posterior technique as described by Sampson et al will lead to complications, and the lateral approach with modifications is still preferable.
With either approach, the advantages of arthroscopic versus open subacromial decompression are evident and include the following:
The disadvantages are the significant learning curve and the increased equipment needs of the arthroscopic procedure. Determination of the amount of bone resection especially with the lateral approach, may be more difficult than with open techniques. Complications, if encountered, may be harder to deal with arthroscopically than with an open procedure.
Impingement is a nonspecific clinical syndrome with a number of different underlying etiologies. Accurate diagnosis is imperative to ensure appropriate nonoperative or surgical treatment. Patients complaining of pain with overhead activities are differentiated into one of the following categories:
Neer1 introduced the concept of extrinsic impingement of the anterior acromion, coracoacromial arch, and the acromioclavicular joint on the underlying rotator cuff and biceps tendon. He also emphasized that forward flexion of the arm is the dominant functional position and that anterior decompression, not lateral acromionectomy, is the appropriate operative approach for significant cuff degeneration. His impingement sign is performed with the patient seated in front of the examiner, who stabilizes the scapula as the arm is elevated slightly lateral to the midline to impinge the tuberosity against the acromion (Fig. 2.1.1).
Pain thus produced is eliminated by injecting 10 cc of 1% Xylocaine into the subacromial bursa beneath the anterior acromion (impingement injection test) to confirm the diagnosis. Hawkins and Kennedy4 described a second impingement sign in which the arm is flexed forward 90 degrees and then forcibly internally rotated, jamming the supraspinatus tendon against the anterior edge of the coracoacromial ligament to produce pain.
Patients with primary extrinsic impingement are usually in an older age group or have a bony architecture with an anterior acromial hook or spur that presses directly on the cuff and biceps with forward elevation of the arm. There is also a younger subgroup of overhead athletes who have benign bony anatomy but have a prominent or hypertrophied anterolateral band of the coracoacromial ligament.12 This produces an extrinsic irritation of the underlying bursa and cuff and occasionally a snap or click. Both of these types of patients have the most predictable operative success with arthroscopic subacromial decompression or coracoacromial ligament resection when conservative treatment has failed.
The concept of secondary impingement originates with Codman,13 who proposed an intrinsic tendinous degeneration as the essential lesion in rotator cuff disease. The micro vascular studies by Rathbun and McNab,14 Moseley and Goldie,15 and Rothman and Parke16 support this concept. This vascular compromise results in tissue devitalization characterized as “angiofibroblastic hyperplasia” by Nirschl.17 The subsequent pain and weakness of the supraspinatus compromises its function as a humeral head depressor and allows the upward humeral migration forces of the deltoid to dominate, producing a secondary impingement of the cuff into the acromion.
F. Jobe et al18 enlarged this concept to include patients with underlying anterior glenohumeral ligament instability. As the humeral head subluxes anteriorly, the cuff is secondarily compressed against the coracoacromial arch.
Secondary impingement is more prevalent in a younger patient population actively involved in sports activities that entail overhead arm motion, and should be suspected when the bony architecture is unremarkable. The subluxation-relocation test, as described by Jobe et al18 is helpful in differentiating secondary causes of impingement (Fig. 2.1.3). With the arm abducted 90° and externally rotated, an anterior force is applied by the examiner’s hand on the posterior aspect of the humeral head. This accentuates the impingement pain in an unstable shoulder as the head and overlying cuff drive into the anterior edge of the acromial arch (subluxatiori). Conversely, posterior pressure on the head alleviates the impingement discomfort (relocation).
Walsch et al19 and C. Jobe20 more recently have described another variety of impingement noted in overhead athletes that occurs when the arm is maximally externally rotated while abducted and extended (such as in the cocking phase of throwing). In this position the posterior superior articular surface fibers of the supraspinatus are placed under tension and sheer but are also compressed between the humeral head and adjacent glenoid rim, resulting in posterior superior synovitis and partial under-surface tears. Whether or not any underlying instability is a factor in this compression is still unresolved. While easily confused with primary or secondary anterior impingement, careful examination usually demonstrates pain more at the posterior-superior aspect of the rotator cuff with the arm abducted and externally rotated and extended, in contrast to the impingement positions of Neer and Hawkins. This apprehension position, although painful in this syndrome, does not elicit the usual anxiety found in patients with instability. However, there still may be a reduction of pain with the relocation maneuver of the subluxation-relocation test described by Jobe.
Gerber et al21 have described this type of anterior impingement between the humeral head and the coracoid process secondary to traumatic, iatrogenic, or idiopathic causes. Whatever the underlying etiology, the tip of the coracoid is positioned more lateral than normal, and as the arm is brought into forward flexion there is a compression of the rotator cuff between the humeral head and the tip of the coracoid. This produces pain with Neer’s forward flexion test, but it occurs usually between 80 and 130 degrees of flexion rather than at full flexion. Also Hawkin’s flexion and internal rotation test is consistently positive, but the pain is lower and more anterior than with superior impingement. The patient also demonstrates decreased horizontal adduction with pain similar to that found with acromioclavicular (AC) disease (Fig. 2.1.4), but the pain is again more at the tip of the coracoid and not at the AC joint.
Gartsman22 coined the term pseudoimpingement syndrome for patients who demonstrated clinical history and physical findings of anterior superior impingement but in whom impingement was due to a lack of full external rotation. This restriction in range of motion does not allow the humerus to rotate externally with elevation, and the rotator cuff is compressed between the greater tuberosity and the acromion when the arm is elevated. This problem is easily confused with primary extrinsic compression but routinely resolves with therapy directed at regaining the lost external rotation.
Knowledge of the coracoacromial anatomy is crucial both for diagnostic accuracy and operative facility, and the avoidance of complications.
The bony architecture is composed of the acromion, the AC joint, the coracoid process, and the greater humeral tuberosity. The shape of the acromion and contour of its undersurface is best evaluated with Neer’s supraspinatus outlet view (Fig. 2.1.5). Bigliani et al23 described three distinct acromial shapes: type 1, flat; type 2, curved; and type 3, hooked. They found an increased correlation between the type III hooked acromion and underlying full-thickness rotator cuff tears (69.5% for type 3 and 3% for type 1). This radiographic view is also valuable in determining the overall slope and thickness of the acromion, and in predetermining those cases where the cutting block technique of acromioplasty would be inappropriate.
Rockwood and Lyons24 have described a modified anteroposterior (AP) view of the shoulder for differentiating the hooked acromion. This x-ray involves angulating the beam 30 degrees caudad to accentuate the anterior acromial protruberance (Fig. 2.1.6). Although this view is helpful in terms of diagnosis, it is not particularly useful in terms of preoperative planning or determining whether to use a lateral or a posterior approach for the acromioplasty.
The AC joint borders the coracoacromial space medially. As it degenerates, it may play an active role in the extrinsic impingement process. Osteophytic overgrowth on the undersurface of the distal clavicle and medial acromion can impinge on the underlying rotator cuff. The pain of an arthritic or osteolytic joint can also mimic that of anterior impingement. Careful preoperative evaluation is necessary to avoid residual pain at the AC joint after decompression.
The coracoid process forms the anterior border of the subacromial space. It may be enlarged, fractured, or iatrogenically altered, such as occurs with a laterally positioned Bristow transfer of the coracoid tip onto the anterior glenoid rim. Fractures of the coracoid can occur with the recoil of a rifle into the shoulder in hunters. A posterior opening wedge osteotomy for instability also effectively lateralizes the coracoid tip relative to the humeral head. These changes, which can be associated with anterior subcoracoid impingement, are best noted on axillary view x-rays or a computed axial tomography (CAT) scan with the arm flexed 90 degrees and internally rotated.
The greater tuberosity of the humerus forms the floor of the coracoacromial space. It is important to note its size and shape, any osteophytic overgrowth, sclerosis, erosion, or cysts. It is best evaluated radiographically with an AP view with the arm in external rotation.
It is important to remember that the subacromial bursa is an anterior structure. It extends from the anterior one-half to one-third of the acromion to just medial to the AC joint to 1 to 2 cm anterior to the acromion and 2 to 3 cm laterally (Fig. 2.1.7). The bursal wall is frequently thickened and troublesome posteriorly, and has been named the “posterior bursal curtain.” This curtain frequently “closes” as one backs the scope posteriorly to get a larger field of view of the subacromial bursa. It is frequently necessary to resect a portion of this structure when performing subacromial surgery.
The anatomy of the coracoacromial ligament is pertinent to the technique of acromioplasty. It attaches to the front and undersurface of the acromion as a thick band and continues around the anterolateral corner to attach to the lateral ridge for a variable distance. Anteriorly the coracoacromial ligament attaches to the anterior inferior edge of the acromion, while the deltoid fascia attaches more superiorly (Fig. 2.1.8). As the coracoacromial ligament is detached, it falls away easily from the overlying anterior deltoid muscle and fascia. Laterally, however, the coracoacromial ligament blends inextricably with the deltoid muscle fascia along the lateral acromion. Care must be taken not to aggressively detach the fascia or resect too much bone laterally, as this may result in a deltoid detachment.
Gallino et al26 found that the CA ligament has a variable thickness of insertion on the undersurface of the acromion, ranging from 2 to 5.6 mm. Those patients with excessively thickened ligaments would be the ones most likely to have anterior functional stenosis and/or snapping, as described by O’Boyle et al,12 and benefit from anterolateral band resection.
Edelson and Luchs25 and others have noted various degrees of transformation of the coracoacromial ligament into bone at its acromial insertion. Gartsman22labeled this phenomenon “anterior acromial protruberance.” Rockwood5 in his open technique recommends resecting 8 to 10 mm of full-thickness anterior bone and then reattaching the deltoid fascia. This technique of full-thickness anterior bone resection back to the level of the AC joint has insinuated itself into some authors’ description of subacromial decompression.26 For the most part the anterior acromial protruberance is really an inferior extension of calcification into the coracoacromial ligament insertion. One does not need to resect full-thickness acromial bone anteriorly to remove it, and in fact great care should be taken not to resect too much superior anterior bone, as this may detach the anterior deltoid fascia producing an operative disaster. The best radiographic views for determining the amount of anterior acromial protruberance are the axillary view and the supraspinatus outlet view (Fig. 2.1.9). The axillary x-ray is also an excellent view for evaluation of the AC joint, particularly for picking up posterior AC arthritis that may be missed on a routine AP view.
The history is important. Pain with the cocking and acceleration phase of throwing is most likely secondary to an underlying instability or posterior superior impingement. Nocturnal and rest pain is often indicative of a rotator cuff, whereas patients with cuff tendinitis develop pain with progressive activity.27 Other causes of shoulder pain such as scapular thoracic bursitis, suprascapular nerve syndrome, cervical radiculopathy, and referred pain from the gallbladder, liver, lung, or heart also need to be differentiated.
The clinical signs and x-rays noted previously are the most valuable in making a diagnosis of impingement. Concomitant rotator cuff disease or AC joint disease can be evaluated with both an arthrogram or magnetic resonance imaging (MRI). The arthrogram may be more accurate in determining full-thickness rotator cuff tears but less sensitive in picking up partial-thickness lesions or intratendinous pathology. Isolated AC joint injection and/or bone scan may be helpful in differentiating AC joint versus sub-acromial disease. It is important to know the status of the AC joint prior to arthroscopic decompression so that residual pathology in this location is not left unattended.
Conservative care should be diligent and prolonged. The goal is to diminish the inflammation in the tissues and then regain full range of motion and full strength in the scapular stabilizers and rotator cuff to balance the deltoid force couple. This is accomplished with rest, hot and cold modalities, massage, nonsteroidal antiinflammatories, and selective injection. Directed physical therapy and home treatment programs are beneficial. Various authors have recommended from 6 to 18 months of conservative care prior to consideration of operative intervention.
Careful preoperative evaluation is necessary to determine the appropriate operative approach and to avoid complications. Outlet and axillary views are the key to evaluating the acromion. The outlet view is utilized to determine the shape of the acromion (type II or type III) and the overall thickness.22,29 On the outlet view, lines are drawn on the undersurface of the acromion—one from the front tip of the acromion to the posterior edge, and a second line along the posterior half of the undersurface of the acromion extending out anteriorly. The distance between these two lines at the anterior margin approximates the amount of undersurface anterior bone that will be resected (Fig. 2.1.10).
The axillary view is used to determine the shape of the acromion (cobra versus square tipped) and whether there is any anterior acromial protruberance. If present, this protuberance will need to be resected at the time of coracoacromial ligament release.
If on the outlet view one notes a very thin or curved acromion, the cutting block line on the undersurface of the posterior half of the acromion may actually exit the superior aspect of the acromion, taking off too much anterior bone (Fig. 2.1.11). In these cases, the cutting block technique, as described by Sampson et al,11 would be inappropriate. Instead, the lateral approach (described below) would be more applicable, removing just a small anterior hook and not producing a type I flat acromion.
Poor visualization in the subacromial space is one of the more frustrating aspects of either approach and is usually secondary to either excessive bleeding or inadequate debridement of the subacromial space. Use of electrocautery is strongly recommended. Other strategies to control bleeding during arthroscopic subacromial decompression include the following:
I perform the procedure in an outpatient setting with the patient in the lateral decubitus position. I use general anesthesia. I don’t routinely use an interscalene nerve block, but this may ensure better postoperative pain control. The procedure may also be done in a beach-chair position with regional anesthesia as per surgeon preference.
The table is turned approximately 100 to 110 degrees from the anesthesiologist, who is then situated at the patient’s abdomen. Long anesthesia tubing is required. The TV monitor tower with contained video equipment is positioned directly anterior to the patient’s head and chest. The shoulder holder is attached to the operating table on the anterior side of the body near the foot. The inflow pump is positioned so that it can be observed by the surgeon during the procedure.
The patient is positioned in the modified lateral decubitus position as described by Gross and Fitzgibbons.30 This position rolls the patient back 25 to 30 degrees, placing the glenoid orientation parallel to the floor (Fig. 2.1.13). The patient is placed in the beanbag with the U position toward the head and the tails extending to the superior-anterior and posterior chest cephad to the axilla for support. The shoulder is isolated with large plastic U drapes, and traction is applied to the patient’s arm. An axillary roll and appropriate head support is utilized. The arm is positioned at approximately 30 degrees of abduction and 10 degrees of flexion with 7 to 15 pounds of traction applied depending on the patient’s size and muscularity. A second dual-traction apparatus may be applied if a stabilization procedure needs to be performed.
The anatomy of the shoulder is outlined with a marking pen prior to the operative procedure and the portals marked. The glenohumeral joint is then examined completely from both a posterior and a high anterior portal, established inside out at the superior aspect of the rotator interval. This will later be the anterior portal for the subacromial bursoscopy. Any pathology within the glenohumeral joint is appropriately addressed.
Partial undersurface or small complete rotator cuff tears are frequently marked with a tag suture placed through an 18-gauge needle introduced from superiorly into the joint and retrieved out the anterior portal (Fig. 2.1.14). This suture marker is beneficial later when subacromial bursoscopy is performed, as it provides a quick reference to the questionable cuff area from the superior view. The scope is then removed from the glenohumeral joint and through the same posterior skin portal, redirected at a 10-degree caudad angle to the acromion into the subacromial bursa and far enough anteriorly to enter the chamber. If the bursa is easily entered and distended, then the inflow is brought in at the scope with a pump and a lateral portal is then made on the basis of an accurately placed 18-gauge needle.
If the bursa is significantly inflamed or not easily distended, with poor visualization, then the scope trocar and sheath is brought directly out anteriorly just lateral to the coracoacromial ligament to exit from the previously made high anterior skin portal. The outflow cannula is then placed on the tip of the trocar and pushed back into the subacromial space so that it lies under the anterior half of the acromion. The sheath is separated slightly, the scope is inserted into the posterior cannula, and flow and visualization are established. A lateral portal is then directed with an 18-gauge needle.
The bursa is then viewed from posteriorly and debrided from the lateral portal until good visualization is established. Any suspicious areas of the rotator cuff that may have been previously identified with a suture marker are debrided and examined from both the posterior portal and the lateral portal.
Preoperatively I will have decided whether I am going to use a modified lateral approach or a cutting-block approach for the decompression. If the patient has a thin curved acromion and a lateral approach is appropriate, I place my lateral portal 3.5 to 4 cm lateral to the acromion and about midway between the midportion of the acromion and the anterolateral corner. I make sure with an 18-gauge needle that I can get the shaver along the anterior-inferior edge of the acromion and a short distance down the anterolateral side, and that it can be directed slightly upward at the acromion for ease in burring and shaving.
The undersurface of the anterior half of the acromion is then debrided with an aggressive shaver and/or a cautery ablation system (Fig. 2.1.15). Care should be taken with either instrument to stay on the undersurface of the bone and not pop off anteriorly or laterally into the deltoid fibers, which are very vascular. The anterolateral corner of the acromion is identified with an 18-gauge needle directed from superiorly, and the debridement is started at this point and progresses medially toward the AC joint and also posteriorly.
From the preoperative planning, the amount of bone to be resected is known, as is the diameter of the burr. Starting at the anterolateral corner, the appropriate amount of anterior hook is resected from anterior to medial. Care is taken not to remove full-thickness bone anteriorly and thereby detach the anterior deltoid fascia. This cannot be subsequently repaired as in open operative procedures. After the anterior bone is resected from lateral to medial, tapering of the remaining posterior bone is then accomplished from anterior to posterior to the midportion of the clavicle, or the scope can be placed laterally and the shaver introduced posteriorly to taper from posterior to anterior. Because of the thin and curved nature of the acromion, the goal is not to produce a completely flat undersurface but to perform a smooth and even taper (Fig. 2.1.16). Whether one tapers from anterior to posterior or posterior to anterior, the scope is always placed laterally to evaluate the decompression in two planes.
If there is no evidence of degenerative disease of the AC joint and no inferior osteophytes, I do not take the decompression into the joint or bevel it. If inferior osteophytes are present, then the undersurface of the AC joint is exposed and the osteophytes removed. Manual pressure from above will then deliver a portion of the distal clavicle to view and if it is noticeably arthritic, an arthroscopic distal clavicle resection can be performed. If the articular cartilage looks healthy, then the beveling alone would be performed.
Following adequate decompression, the pump pressure is reduced and hemostasis is obtained with the electrocautery unit. The subacromial space is then instilled with 10 cc of 0.25% bupivacaine with epinephrine and then 1 to 2 cc in each incision. The portals are closed with 4-0 nylon and a sterile dressing is applied.
Immediate postoperative motion is allowed and encouraged. No sling is utilized. On the first postoperative day, passive and active motion is encouraged to avoid the possibility of developing an adhesive capsulitis or captured shoulder, as described by Gross’s group.31 Patients are allowed to return to sedentary work as soon as possible. Heavy manual labor usually requires a slower progression and may take from 6 to 12 weeks.
I routinely utilize a two-portal cutting-block technique as described by Sampson et al.11 Although I orient the acromion on the top of the screen when I am in either the posterior or lateral portal, the principles of the procedure still apply. I have found this technique to be considerably more reproduceable and reliable than the traditional lateral approach as described herein and by Ellman8 and utilize it for at least 95% of my subacromial decompressions. On the rare occasions where a thin, broad, and curved acromion is encountered, then the cutting-block technique is inappropriate. The lateral approach as described above is still utilized with success.
1. Neer CS. Anterior acromioplasty for chronic impingement syndrome in the shoulder: a preliminary report. J Bone Joint Surg 1972;54A:41-50.
2. Neer CS. Impingement lesions. Clin Orthop 1983;173:70-77.
3. Hawkins RJ, Brock RM, Abrams JS, et al. Acromioplasty for impingement with an intact rotator cuff. J Bone Joint Surg 1986;70B:795-797.
4. Hawkins RJ, Kennedy JC. Impingement syndrome in athletes. Am J Spans Med 1980;8:151-158.
5. Rockwood CA Jr. Surgical treatment of the shoulder impingement syndrome: a modification of the Neer anterior acromioplasty in 71 shoulders. Orthop Trans 1990;14:251.
6. Bigliani LU, Delessandro DF, Duralde XA, et al. Anterior acromioplasty for subacromial impingement in patients younger than 40 years of age. Clin Orthop 1989;246:111-116.
7. Johnson LL. Shoulder arthroscopy. In: Johnson LL, ed. Arthroscopic surgery: principles and practice. St. Louis: CV Mosby, 1986;1371-1379.
8. Ellman H. Arthroscopic subacromial decompression: analysis of one to three year results. Arthroscopy 1987;3:173-181.
9. Esch J, Ozerkis LR, Helgager JA, et al. Arthroscopic subacromial decompression: results according to the degree of rotator cuff tear. Arthroscopy 1988;4:241-249.
10. Paulos LE, Franklin JL. Arthroscopic shoulder decompression development and application—five year experience. Am J Sports Med 1990; 18:235-244.
11. Sampson TG, Nisbet JK, Glick JM. decision acromioplasty in arthroscopic subacromial decompression of the shoulder. Arthroscopy 1991;7:301-307.
12. O’Boyle M, Newton PM, Arroyo JS, et al. Arthroscopic resection of the anterolateral and of the coracoacrornial ligament for impingement in the overhead athlete. Paper presented at the 16th Annual Meeting of the Arthroscopy Association of North America, San Diego, California, April 1997.
13. Codman EA. Rupture of the supraspinatus tendon and other lesions in or about the subacromial bursa. In: The shoulder. Boston: Thomas Todd, 1934;73-75.
14. Rathbun JB, McNab I. The microvascular pattern of the rotator cuff. J Bone Joint Surg 1970;52B:540-553.
15. Moseley HF, Goldie I. The arterial pattern of the rotator cuff on the shoulder. J Bone Joint Surg 1963;45B:780-789.
16. Rothman RH, Parke WW. The vascular anatomy of the rotator cuff. Clin Orthop 1965;41:176-186.
17. Nirschl RP. Rotator cuff tendinitis: basic concepts of pathoetiology. Instr Course Lect 1989;38:439-445.
18. Hibe FW, Kvitne RS, Giangarra CE. Shoulder pain in the overhand or throwing athlete: the relationship of anterior instability and rotator cuff impingement. Orthop Rev 1989;18:963-975.
19. Walsch G, Boylau P, Noel E, et al. Impingement of the deep surface of the supraspinatus tendon on the posterior superior glenoid rim: an arthroscopic study. J Shoulder Elbow Surg 1992; 1:238-245.
20. Jobe CM. Posterior superior glenoid impingement: expanded spectrum Arthroscopy. 1995;11:530-536.
21. Gerber C, Terier F, Ganz R, The role of the coracoid process in chronic impingement syndrome. J Bone Joint Surg 1985; 678:703-708.
22. Gartsman GM. Arthroscopic acromioplasty for lesions of the rotator cuff. J Bone Joint Surg 1990;72A: 169-180.
23. Bigliani LU, Morrison DS, April EW. The morphology of the acromion and its relationship to rotator cuff tears. Orthop Trans 1986;10:216.
24. Rockwood CA Jr, Lyons FR. Shoulder impingement syndrome: diagnosis, radiographic evaluation, and treatment with a modified Neer acromioplasty. J Bone Joint Surg 1993;75A: 409-424.
25. Edelson JG, Luchs J. Aspects of coracoacrornial ligament anatomy of interest to the arthroscopic surgeon. Arthroscopy 1995;11:715-719.
26. Gallino M, Vatiston B, Annaratone G, et al. Coracoacromiol ligament: a comparative arthroscopic and anatomic study. Arthroscopy 1995;ll:564-567.
27. Esch JC, Baker CL. Rotator cuff disease and impingement. In: Whipple TL, ed. Arthroscopic surgery—the shoulder and elbow. Philadelphia: JB Lippincott, 1993:161-163.
28. Rockwood CA Jr. Shoulder function following decompression and irrepairable cuff lesions. Orthop Trans 1984;8:92.
29. Wuh HCK, Snyder SJ. Modified classification of the supraspinatus outlet view based on the configuration and the anatomical thickness of the acromion. Paper presented at the Fifty-Ninth Annual Meeting of the American Academy of Orthopedic Surgeons, Washington, DC, February 1992.
30. Gross RM, Fitzgibbons TC. Shoulder arthroscopy: A modified approach. Arthroscopy 1985;1:156-159.
31. Mormino MA, Gross RM, McCarthy JA. Captured shoulder a complication of rotator cuff surgery. Arthroscopy 1996;12 457-461.