Arthroscopic Management of Frozen Shoulder

Peter N. Chalmers, MD

Seth L. Sherman, MD

Neil Ghodadra, MD

Gregory P. Nicholson, MD

Dr. Nicholson or an immediate family member has received royalties from Innomed and Zimmer; serves as a paid consultant to or is an employee of Zimmer and Tornier; has stock or stock options held in Zimmer; and has received research or institutional support from EBI, Tornier, and Zimmer. None of the following authors or any immediate family member has received anything of value from or has stock or stock options held in a commercial company or institution related directly or indirectly to the subject of this chapter: Dr. Chalmers, Dr. Sherman, and Dr. Ghodadra.

PATIENT SELECTION

Frozen shoulder is defined by loss of range of motion of the shoulder, impairing the patient’s ability to sleep, work, perform activities of daily living, or perform desired recreational activities. Etiology, pathology, natural history, diagnosis, and treatment have been debated.¹^,²^,³^,⁴ Cytokines, myofibroblasts, growth factors, and matrix metalloproteinases have all been implicated, and similarities to both Dupuytren disease and Peyronie disease have been described.³^,⁴ Frozen shoulder may be the end-stage manifestation of several primary conditions, including trauma, surgery, prolonged immobilization, endocrine disorders such as hypothyroidism or diabetes mellitus, and idiopathic causes.⁵ This condition generally affects patients 40 to 60 years of age and occurs in roughly 2% of the population but up to 18% of diabetic patients.¹^,² The natural history of idiopathic frozen shoulder is variable but generally progresses in three stages, each 6 to 9 months in length: (1) inflammation (freezing), (2) fibrosis with disorganization and contracture (frozen), and (3) resolution (thawing).¹^,² Although resolution generally occurs within 2 to 3 years in idiopathic frozen shoulder, some limitation of range of motion of questionable functional significance may be permanent.¹ Because of the combination of inflammation and fibrosis seen on pathology, the terms frozen shoulder and adhesive capsulitis have been used interchangeably.

Frozen shoulder is a clinical diagnosis. The primary maneuver used to diagnose frozen shoulder is the assessment of range of motion before and after the injection of local anesthetic into the glenohumeral joint. If limitation in range of motion resolves after the injection of local anesthetic, other conditions must be considered, but if range of motion limitation persists, then a diagnosis of frozen shoulder is more likely. Evaluation of range of motion after subacromial injection can also be useful to exclude impingement as the primary pathology limiting shoulder mobility. Other conditions, such as rotator cuff tears, acromioclavicular joint pathology, subacromial impingement, and cervical spine pathology, should all be excluded before diagnosing a patient with frozen shoulder.

A variety of treatments have been described for frozen shoulder, including benign neglect, nonsteroidal anti-inflammatory drugs (NSAIDs), oral steroids, steroid injections, home stretching regimens, supervised physical therapy, brisement, manipulation under anesthesia (MUA), arthroscopic synovectomy, arthroscopic subacromial decompression, arthroscopic capsular release, open capsular release, and combinations thereof.²^,⁵^,⁶^,⁷^,⁸^,⁹^,¹⁰ Nonsurgical measures, including NSAIDs (nonsteroidal anti-inflammatory drugs), oral steroids, steroid injections, home stretching, and physical therapy by a therapist with an understanding of frozen shoulder should be pursued for at least 6 weeks to 3 months before surgical intervention. Almost 90% of patients respond to nonsurgical treatment.⁷^,⁸ If patients have had symptoms for 3 to 6 months, had nonsurgical treatment for a known diagnosis of shoulder stiffness for at least 6 to 8 weeks, and still continue to experience pain and decreased range of motion interfering with their ability to sleep, work, complete activities of daily living, or participate in sporting activities as they desire, surgical intervention should be considered.

Arthroscopic capsular release has been demonstrated to decrease pain, improve range of motion, improve function, and possibly shorten the natural history of the disorder.²^,³^,⁵^,¹⁰^,¹¹ Contraindications to arthroscopic capsular release include an indwelling prosthesis or a history of instability. A controversial relative contraindication to arthroscopic capsular release is a history of open surgical intervention. In this circumstance, range of motion limitation may be secondary to scarring between tissue layers, which can be a more technically demanding arthroscopic procedure.

Historically, MUA was performed for these indications and has been shown to be effective in a recent meta-analysis.⁶ Advantages of arthroscopic capsular release over MUA include (1) additional diagnostic information gained by arthroscopy with respect to the subacromial space, the state of the articular cartilage, and other intra-articular pathologies; (2) a more controlled and precise capsular release of the structures selected, instead of the ripping avulsion of the weakest structures produced by MUA;
(3) the ability to perform a simultaneous brisement of the joint via insufflation; (4) the ability to perform concomitant synovial débridement, which has been discussed as the possible mediator of the pathologic process and which was performed in 80% of shoulders in the series reported by one of the authors (G.P.N.)⁵; and (5) the ability to perform subacromial decompression, which was performed in 33% of shoulders in that series.⁵ Additionally, MUA has been associated with proximal humeral fracture, axillary nerve neurapraxias, labral injuries, subscapularis tears, superior labral anterior-to-posterior (SLAP) tears, and articular cartilage injuries.⁹^,¹² Arthroscopic release also has several advantages over open release, including decreased surgical morbidity, decreased length of stay, decreased postoperative pain, and improved ability to address the posterior capsule. Because of these advantages, arthroscopic capsular release has become the standard surgical approach to frozen shoulder.

PREOPERATIVE IMAGING

Although frozen shoulder is a clinical diagnosis, it is also one of exclusion; thus, imaging can be helpful to exclude other pathology. AP, lateral, outlet, and Zanca views of the shoulder can be used to evaluate for glenohumeral arthritis, fracture, locked dislocation, chondrolysis, calcific tendinitis, acromioclavicular arthrosis, abnormal acromial morphology, and other disorders. Shoulder MRI can be useful to evaluate the rotator cuff, biceps tendon, and glenohumeral and coracohumeral ligaments. Shoulder arthrography was traditionally used to document decreased joint volume, but this method is currently only of historical interest.

VIDEO 25.1 Arthroscopic Management of the Frozen Shoulder. Peter N. Chalmers, MD; Seth L. Sherman, MD; Neil Ghodadra, MD; Gregory P. Nicholson, MD (4 min)

View Quicktime Video

Video 25.1

PROCEDURE

Patient/Positioning/Equipment

Capsular release for frozen shoulder is among the more difficult arthroscopic procedures in the shoulder. Contracted capsule can limit both visualization and the mobility of instruments within the joint. The surgeon must be patient and recognize the limited visualization for this procedure without becoming frustrated. In addition to general arthroscopy equipment, an insufflation pump that allows control of flow and pressure, an articulated arm holder, and 90° bipolar electrocautery can greatly assist the surgeon.

In addition to a scalene block with a long-acting local anesthetic or a scalene catheter to allow postoperative pain control, patients also receive general anesthesia to assist in muscular relaxation. Hypotensive anesthesia decreases intraoperative bleeding. Both the beach-chair position and the lateral decubitus position with axial traction have been described. The beach-chair position allows subtle rotational positioning changes and eases conversion to an open procedure (Figure 1). A rolled towel is placed beneath the medial border of the scapula. An articulated arm holder is attached to the operating table and covered with a sterile drape to facilitate finding and maintaining the subtle positioning changes that allow visualization of the joint and the excursion of instruments within it. An experienced assistant aids greatly with arm positioning. The acromion, distal clavicle, and coracoid are marked (Figure 2). After anesthesia has been administered, range of motion is documented, including elevation, external rotation in adduction, external rotation in abduction, and internal rotation. Internal rotation may be sufficiently limited to make measurement in adduction difficult. Internal rotation also should be measured in 40° of abduction in the scapular plane. Manipulation of the joint before arthroscopy causes hemorrhage, making subsequent joint visualization difficult.¹² We do not routinely manipulate before arthroscopy.

Only gold members can continue reading. Log In or Register to continue