Glenohumeral Arthritis

Published on 11/03/2015 by admin

Last modified 22/04/2025

Print this page

This article have been viewed 1500 times

CHAPTER 28 Glenohumeral Arthritis

Stephen C. Weber, Jeffrey I. Kauffman, Carol A. Parise

Initially, arthroscopic management of shoulder arthritis was limited to diagnosis. The pioneering work of Ellman and colleagues ¹ and Cofield² was primarily directed at differentiating arthritis from other shoulder diagnoses, especially in its early stages, where it is often radiographically silent. Treatment of these serendipitously identified lesions was limited to lavage and débridement, with little information as to functional outcomes. The outcome of arthroplasty, especially in younger individuals, has also been better defined. Although showing improvement, Sperling and associates³ have shown that half of all patients who have an arthroplasty for shoulder arthritis who are younger than 50 years have an unsatisfactory outcome, whether or not the glenoid is resurfaced. The poor outcomes and limited durability of arthroplasty in young individuals has led to renewed interest in nonarthroplasty treatment methods for glenohumeral arthritis, especially those techniques that can be performed arthroscopically. Arthroscopic débridement for arthritis now has reasonable midrange results reported, with significant information to relate to the prospective patient in regard to anticipated outcome. Although other techniques have also been reported, follow-up and detailed information about outcomes remain limited.

CLASSIFICATION

Arthritis of the shoulder has traditionally been classified into inflammatory or noninflammatory arthropathies (Table 28-1). Each diagnosis will have its unique presentations and difficulties. The primary unifying factor is the presence or absence of significant synovitis with the inflammatory arthropathies for which arthroscopic synovectomy is recommended. Recent peer-reviewed studies have shown significant benefit for synovectomy and arthroscopic treatment in patients with rheumatoid arthritis, although progression of disease is not universally halted, and maximal benefits were seen in patients with more modest radiographic changes.⁴ Conversely, synovectomy would rarely be indicated in noninflammatory arthropathy.

TABLE 28-1 Classification of Glenohumeral Arthritis

Inflammatory	Noninflammatory
Rheumatoid arthritis	Primary osteoarthritis
Systemic lupus erythematosus	Posttraumatic osteoarthritis
Gout	Septic arthritis
Calcium pyrophosphate deposition	Postinstability arthritis
Psoriatic arthritis	Capsulorrhaphy arthropathy
Ankylosing spondylitis	Charcot arthritis

Gout is uncommon in the shoulder, and rarely requires arthroscopic treatment. Both ankylosing spondylitis and psoriatic arthritis share a common presentation of concomitant acromioclavicular arthritis, with the requirement for partial distal claviculectomy at surgery. Calcium pyrophosphate deposition is a common finding in degenerative arthritis; its significance in terms of altering treatment for the patient who fails conservative management for their arthritis is unclear.

Noninflammatory arthritis rarely presents with significant synovitis. Arthroscopic treatment for these conditions will vary depending on the associated findings and preference of the surgeon. The Outerbridge classification,⁵ developed for the knee, has been widely used to classify the extent of arthritic change in the shoulder (Table 28-2). Recent data have shown that the Outerbridge classification is of significant benefit in predicting the outcome of arthroscopic treatment of the shoulder. Radiographic classification is also of value. Although many classifications are available, that of Samilson and Prieto⁶ has been widely used to assess the outcome of arthroscopic treatment of the arthritic shoulder (Table 28-3). Chondrolysis has been a challenging problem that has been associated with thermal capsulorrhaphy and pain pumps with continuous infusion of local anesthetics. The cause of chondrolysis in general remains unclear; degeneration, inflammation or infection may play a role.⁷

TABLE 28-2 Glenohumeral Joint: Arthroscopic Grading Criteria

Grade	Cartilage
I	Softening or blistering
II	Fissuring or fibrillation
III	Deep ulceration
IV	Exposed bone

TABLE 28-3 Samilson Classification of Shoulder Arthritis

Grade	Description
I	No changes
II	Mild arthrosis, osteophyte <3 mm
III	Moderate arthrosis, osteophyte 3- 7 mm, some joint space irregularity
IV	Severe arthrosis, osteophyte >7 mm, joint space narrowing, sclerosis

From Samilson RL, Prieto V. Dislocation arthropathy of the shoulder. J Bone Joint Surg Am. 1983;65:456-460.

ANATOMY AND PATHOANATOMY

Anatomic considerations are those for shoulder arthroscopy in general. The thickness of articular cartilage is variable in the shoulder. There is a normal bare area on the posterior humerus that can involve as much as 20% of the humeral surface. The glenoid similarly has a bare area similar to the fovea in the hip. The glenoid bare area can be used as a landmark for the anatomic center of the glenoid. Capsular contractures can be responsible for restriction of motion and increased shear forces across the articular cartilage. Glenohumeral osteoarthritis classically presents with anterior capsular contracture causing loss of external rotation and posterior subluxation. With time, posterior glenoid erosion can occur, causing a biconcave glenoid. Issues arise in regard to successful joint entry and movement of instruments, and can present challenges requiring meticulous attention to detail in regard to establishing portals and avoiding instrument breakage.

PHYSICAL EXAMINATION

Findings for arthritis, especially in the early stages, can often be vague and nonspecific, as initially noted by Ellman and coworkers.¹ Arthroscopy and synovial biopsy remain principal mechanisms of diagnosis for early arthritis of the inflammatory or degenerative type. Nonspecific deep tenderness, especially posteriorly, can be a clue. Pain at the extremes of motion may be present. The hallmark of later degenerative glenohumeral osteoarthritis is progressive loss of active and passive external rotation of motion. Crepitus is nonspecific, but can be present in inflammatory and noninflammatory arthritis. Warmth, redness, and swelling are nonspecific findings. Often, differential injections can be very useful. Patients with early glenohumeral arthritis will have a negative subacromial lidocaine impingement test but a positive response to intra-articular lidocaine injection.

DIAGNOSTIC IMAGING

Changes of early arthritis are often difficult to detect on plain radiographs, as reported early on by Ellman and colleagues.¹ Later osteoarthritic changes such as joint space narrowing, posterior subluxation, and the common anteroinferior osteophyte described by many as the goat’s beard osteophyte. Inflammatory arthritis will rarely present with significant osteophytes. Periarticular erosions, however, are common, resulting from chronic synovitis, and vertical migration caused by associated rotator cuff damage can be seen. MRI scanning similarly can be negative in early disease, and standard MRI scanning with or without contrast will often miss significant articular cartilage injuries. The presence of a large effusion on an MRI scan can indicate an active arthritic process, especially if other obvious causes for the effusion are absent. Investigational MRI techniques may improve the accuracy of MRI for articular cartilage injuries.

TREATMENT OPTIONS

Nonoperative Treatment

Nonoperative treatment remains the mainstay of the management of the arthritic shoulder. Because all treatment would require permanent activity modification in the postoperative period, preoperative activity modification would seem to be in order. Heavy weightlifting, impact activities, and manual labor would at best be expected to eliminate the long-term benefit of any operative treatment, and honest counseling about reasonable activities with the arthritic shoulder is necessary. Anti-inflammatory medications can offer symptomatic relief. Gentle physiotherapy can bring significant improvement in range of motion. Steroid injection can also be of benefit in inflammatory and noninflammatory conditions. Viscosupplementation is not currently approved for the shoulder by the U.S. Food and Drug Administration (FDA) but has shown benefits in the knee, and might offer similar short-term benefits in the shoulder.⁸ Operative treatment should only be considered for those individuals who have failed nonoperative treatment.

Arthroscopic Techniques

Diagnostic Arthroscopy

Diagnostic arthroscopy for arthritis can be performed in the lateral decubitus or beach chair position, although our preference is for lateral decubitus position. Standard portal placement is usually adequate, although placement accuracy is even more important because of the constricted nature of the joint. Joints with extensive contractures will need releases anyway, and these releases should be performed sequentially early to allow safe evaluation of the entire joint to minimize risk of instrumentation and iatrogenic damage to the articular surfaces. Arthroscopic visualization and synovial biopsy can be the mainstay of diagnosis for many arthridites.

Débridement and Lavage

Débridement and lavage have been widely used in arthroscopic treatment of arthritis of the shoulder, much as with arthritis of the knee.⁹ Results are summarized in Table 28-4.¹^,¹⁰^–¹⁷ Several of these studies now have follow-up of at least 60 months. Almost all studies have shown good short-term improvement with minimal complications; long-term results have been positive as well, but generally not as encouraging as the short-term results. In 1997, Norris and Green¹⁸ presented a series of failed arthroscopic débridements that has been widely quoted as emphasizing poor outcomes. This series, however, was limited to patients who presented to these surgeons with failed débridement requesting arthroplasty, and introduces a significant selection bias. Results of our study¹⁰ are summarized in Figures 28-1 to 28-3. Although the early results are encouraging, there is significant degradation of the results with longer term follow-up (see Fig. 28-1). The radiographic score based on the Samilson classification was highly correlated with function at final follow-up (see Fig. 28-2). Although the results tend to diminish with time, only 20% of the patients required arthroplasty at a mean follow-up of 60 months, indicating that débridement appears to be able to defer the need for arthroplasty in a large number of patients (see Fig. 28-3). This is not insignificant in view of the risk of precocious loosening of arthroplasty in younger individuals. Higher grade articular lesions and more severe radiographic changes are associated with inferior long-term results, consistent with similar treatment in the knee.⁹ Techniques vary among studies, but most involve lavage, mechanical smoothing of the articular surfaces, and removal of any loose bodies (Fig. 28-4). Capsular release to remove the standard internal rotation contracture has shown good results.¹⁹ This can be done to correct preoperative loss of motion. Release of the anterior capsule will improve external rotation, and posterior capsular release will improve internal rotation. Release is best done sequentially, checking the gains in motion with each sequential release until full motion is recovered. This prevents overrelease and potential instability. Some have suggested removal of the anteroinferior osteophyte,²⁰ but the risk of iatrogenic injury to the axillary nerve has tempered the enthusiasm for this step. Glenoplasty to restore the glenoid to a concave configuration in patients in whom posterior subluxation has resulted in a biconcave glenoid has been discussed but is also controversial.²¹ Concomitant acromioplasty and acromioclavicular (AC) resection has been shown to provide good outcomes, despite the presence of arthritis.¹²^,²² Recently, Kerr and McCarty²³ have suggested that those patients with lesions of both the glenoid and humerus fare worse with débridement than those with unipolar lesions. Moderate improvement in symptoms and low complication rates favor the use of débridement and acromioplasty as the initial treatment in the management of mild to moderate glenohumeral arthritis. However, the older patient with Samilson grade III or IV changes on radiographs may be better managed with arthroplasty, because their results with arthroplasty tend to be superior to those in younger patients.

TABLE 28-4 Outcomes of Studies of Arthroscopic Treatment of Shoulder Arthritis

FIGURE 28-1 Preoperative (Pre-op), early preoperative (p-op), and final UCLA scores. Note that early results decrease significantly at final follow-up of 60 months (mean).

(Weber SC, Kauffman JI: Arthroscopic débridement for glenohumeral osteoarthritis: long-term follow-up. Paper presented at: AAOS 72nd Annual Meeting; February 23-27, 2005; Washington, DC.)

FIGURE 28-2 Pearson correlation coefficient, UCLA final score versus Samilson radiographic score. Pearson correlation coefficient = 0.430; P < .02.

(Weber SC, Kauffman JI: Arthroscopic débridement for glenohumeral osteoarthritis: long-term follow-up. Paper presented at: AAOS 72nd Annual Meeting; February 23-27, 2005; Washington, DC.)

FIGURE 28-3 Survivorship curve—progression of débrided shoulders to total shoulder replacement at final follow-up.

(Weber SC, Kauffman JI: Arthroscopic débridement for glenohumeral osteoarthritis: long-term follow-up. Paper presented at: AAOS 72nd Annual Meeting; February 23-27, 2005; Washington, DC.)

FIGURE 28-4 Grade IV articular cartilage lesion of the humerus after débridement and lavage.

Marrow Stimulation Techniques

The results of marrow stimulation techniques such as microfracture have shown favorable results in the knee, but remain largely anecdotal in the shoulder (Fig. 28-5). McCarty and Cole have offered an excellent summary of the data available.²⁴ Microfracture has been described in the knee as an open procedure combined with a periosteal flap coverage. Results in this study were generally good; at mean 32-month follow-up, Constant scores increased significantly and osseous integration was proven in two second-look operations.²⁵ It seems reasonable that microfracture of focal lesions of the humerus or glenoid would offer improved outcomes over simple débridement, but data to support this assertion are lacking.

FIGURE 28-5 Microfracture of a grade IV articular cartilage lesion of the glenoid.

Osteochondral Autograft Transfer

Osteochondral autograft transfer has shown good preliminary results in the shoulder, although donor site morbidity from the knee was the most significant complication. Humeral lesions are generally more amenable to osteochondral autograft transfer because of the ability to rotate the humerus to allow perpendicular access to the articular surface. Described as an open procedure by Schiebel and Bartl,²⁶ there would be no reason that these procedures could not be performed arthroscopically, as in the knee. The role of osteochondral autograft transfer again is elusive, but probably would be limited to those patients with focal lesions who fail more traditional, established techniques, such as débridement.

Autogenous Chondrocyte Transfer

Autogenous chondrocyte transfer has been widely used in the knee to manage Outerbridge grade IV lesions. One might anticipate as good or better results in the shoulder, because weight-bearing would not be an issue. Although implantation is not an actual arthroscopic procedure, Romeo and associates ²⁷ have described the successful use of autogenous chondrocyte transfer in a patient with large lesions on the anterosuperior humerus created by chondrolysis secondary to thermal capsullorrhaphy. The lesions were exposed through a deltopectoral approach, and the periosteal graft was taken from the tibia.

Arthroscopic Interposition Arthroplasty

Preliminary results of interposition arthroplasty for the glenoid in association with prosthetic humeral resurfacing have been encouraging.²⁸ These results have led to attempts to perform interposition arthroplasty without prosthetic replacement arthroscopically. Huijsmans and coworkers’²⁹ preliminary study used an acellular allograft skin-derived collagen matrix. Of 11 patients, 6 had 90% improvement, with complete resolution of pain in the short term (8- to 20- month follow-up), 5 showed no improvement, and 1 was converted to a hemiarthroplasty. Similar results have been presented in the peer-reviewed literature with interposition arthroplasty using a lateral meniscus³⁰ and also an acellular orthobiologic xenograft.³¹ Most of the studies had short-term follow-up, with limited outcome data. Clearly, the advantages of avoiding prosthetic replacement in the younger patient are significant, but long-term data on which to make recommendations to pursue these techniques is lacking, and the procedures are technically challenging. Complications in these series were minimal, however, and did not appear to complicate subsequent arthroplasty significantly if required.

PEARLS& PITFALLS

1. Outerbridge grade III and IV lesions often present as incidental findings in the course of arthroscopic evaluation of patients with indications for other procedures, such as acromioplasty.

• These patients should be managed based on their preoperative workup, with débridement and lavage of their articular cartilage injuries.

• Long-term follow-up of these patients, with no radiographic changes, has shown good results, with few problems.

2. Management of patients with obvious radiographic changes of glenohumeral arthritis is more problematic.

• It may be best for older patients with a sedentary lifestyle to defer treatment until they are ready for arthroplasty; the results are generally good, and the risks of revision surgery are low.

• Management of the younger active patient with Samilson grade III or IV radiographic changes is more difficult; arthroplasty is not often successful, and will almost certainly require revision within the patient’s lifetime.

• Given the good short-term results of débridement and lavage, this probably represents the first step for patients who fail initial conservative management.

3. Capsular release can also be beneficial in restoring glenohumeral centering and improving external rotation.

• Patients with focal lesions can be considered for microfracture, because this adds little to the morbidity of the procedure and may be of benefit.

4. There are few long-term studies on which to make a decision for patients who fail this algorithm.

• For patients who do not want to consider arthroplasty, autogenous chondrocyte transfer or interposition arthroplasty have shown preliminary encouraging results and at least seem not to compromise the ability to perform a subsequent arthroplasty if necessary.