HEMIARTHROPLASTY VERSUS TOTAL SHOULDER ARTHROPLASTY: INDICATIONS FOR GLENOID RESURFACING

Indications for glenoid resurfacing in unconstrained shoulder arthroplasty are a much debated topic. We currently favor total shoulder arthroplasty because the results are superior and the complication rate is equal to or lower than that seen in hemiarthroplasty. Our decision to perform glenoid resurfacing or not is detailed according to diagnosis in the following sections.

Two major requirements exist for performance of glenoid resurfacing in unconstrained shoulder arthroplasty: the glenoid bone must be sufficient to allow implantation of components (judged by preoperative imaging studies; see Chapter 7), and the anterior and posterior rotator cuff must be functioning (judged by clinical examination and preoperative imaging studies; see Chapter 7).¹ Lack of either of these requirements represents an absolute contraindication to glenoid resurfacing. A relative contraindication to glenoid resurfacing is young patient age. A “safe” age at which to implant a polyethylene glenoid component has not been established. Concerns of polyethylene wear are heightened in younger patients because of their residual life expectancy. In general, we favor biologic glenoid resurfacing in patients younger than 40 years in whom total shoulder arthroplasty would otherwise be indicated.

PRIMARY OSTEOARTHRITIS

Primary osteoarthritis was initially described by Neer and is the most common single indication for shoulder arthroplasty in our practice.² In a large multicenter study, primary osteoarthritis was the underlying cause in half of the primary shoulder arthroplasties performed.³

Clinical Findings

Clinical findings in patients with primary osteoarthritis include glenohumeral crepitus and stiffness. Rota-tor cuff testing may be normal or compromised by pain.

Imaging Findings

Plain radiography demonstrates loss of the normal glenohumeral joint space. Humeral head osteophytes are usually present and may be large (Fig. 6-1). Loose bodies may be apparent on plain radiography, especially in the subscapularis recess (Fig. 6-2).

Figure 6-1 Radiograph of primary osteoarthritis showing the typical loss of the glenohumeral joint space and the presence of large humeral osteophytes.

Figure 6-2 Large loose body in the subscapularis fossa in a patient with primary osteoarthritis.

Secondary imaging studies (computed tomographic arthrography, magnetic resonance imaging) will show the “classic” posterior glenoid erosion with biconcavity in only 20% of cases (Fig. 6-3).³ Approximately half of patients with primary osteoarthritis will have the humeral head centered within the glenoid, and another 25% will demonstrate posterior subluxation without osseous erosion (Fig. 6-4).³ Less than 5% of patients with primary osteoarthritis will demonstrate a dysplastic-appearing glenoid morphology(Fig. 6-5).³

Figure 6-3 Computed tomography demonstrating the “classic” biconcave glenoid seen in primary osteoarthritis.

Figure 6-4 Computed tomography showing the more common posterior humeral head subluxation without the osseous glenoid erosion seen in primary osteoarthritis.

Figure 6-5 Dysplastic glenoid morphology seen rarely in primary osteoarthritis. Note the excessive glenoid retroversion present with a relatively well centered humeral head.

Seven percent of patients with primary osteoarthritis have a full-thickness rotator cuff tear limited to the supraspinatus, and an additional 7% have a partial-thickness rotator cuff tear.⁴ Moreover, moderate to severe fatty infiltration of the infraspinatus or subscapularis (or both) occurs in approximately 20% of patients with primary osteoarthritis.⁴

Special Considerations

We perform total shoulder arthroplasty in nearly all cases of primary osteoarthritis because it has been shown to be superior to hemiarthroplasty without an increased risk of complications or reoperations.^5,⁶ The only patients with primary osteoarthritis in whom we perform hemiarthroplasty are those with insufficient glenoid bone or anterior or posterior rotator cuff insufficiency. In patients with rotator cuff insufficiency, we usually opt for a reverse-design prosthesis instead of a hemiarthroplasty (see Section Four).

RHEUMATOID ARTHRITIS

Rheumatoid arthritis is the most common inflammatory joint disease. Shoulder manifestations develop in 60% to 90% of patients as their disease progresses. In a large multicenter study, rheumatoid arthritis was the underlying cause in 12% of the primary shoulder arthroplasties performed.⁷

Clinical Findings

Clinical findings in patients with rheumatoid arthritis include glenohumeral crepitus and stiffness. Rotator cuff testing may be normal, compromised by pain, or compromised by a rotator cuff tear.

Imaging Findings

Plain radiography demonstrates loss of the normal glenohumeral joint space. Humeral head osteophytes are rarely present (Fig. 6-6). The humeral head may be centered or statically migrated, depending on the condition of the rotator cuff.

Figure 6-6 Radiograph of rheumatoid arthritis showing the typical loss of the glenohumeral joint space and a paucity of humeral osteophytes.

Secondary imaging studies (computed tomographic arthrography, magnetic resonance imaging) may show protrusio-type glenoid morphology (Fig. 6-7). Eight percent of patients with rheumatoid arthritis have a full-thickness rotator cuff tear limited to the supraspinatus, and an additional 9% have a partial-thickness supraspinatus tear.⁸ Twelve percent of patients with rheumatoid arthritis have massive rotator cuff tears involving the supraspinatus and infraspinatus tendons.⁸ Additionally, moderate to severe fatty infiltration of the infraspinatus or the subscapularis (or both) occurs in approximately 45% of patients with rheumatoid arthritis.⁸

Figure 6-7 Computed tomogram demonstrating a protrusio-type glenoid morphology seen in rheumatoid arthritis.

Special Considerations

We perform total shoulder arthroplasty in nearly all cases of rheumatoid arthritis because it has been shown to be superior to hemiarthroplasty.⁹ The only patients with rheumatoid arthritis in whom we perform hemiarthroplasty are those with insufficient glenoid bone or anterior or posterior rotator cuff insufficiency. In patients with rotator cuff insufficiency, we usually opt for a reverse-design prosthesis instead of a hemiarthroplasty, provided the patient does not have severe osteopenia (see Section Four).

A rare subtype of rheumatoid arthritis that rarely requires shoulder arthroplasty but merits special consideration is juvenile-onset rheumatoid arthritis. In our limited experience, these patients tend to have severe preoperative stiffness that may require more soft tissue releases than normal, including release of the entire pectoralis major. Additionally, the humerus and glenoid may be exceptionally small and thus necessitate the use of custom-manufactured implants.

OTHER INFLAMMATORY ARTHROPATHIES

Other inflammatory arthropathies affecting the glenohumeral joint sufficiently to require shoulder arthroplasty are rare and include Paget’s disease, ankylosing spondylitis, psoriatic arthritis, systemic lupus erythematosus, scleroderma, and polymyalgia rheumatica. Unfortunately, because of the relative rarity of these conditions, little information is available regarding special considerations for these diseases. When any of these conditions evolve to complete loss of articular cartilage and fail reasonable nonoperative treatment, shoulder arthroplasty may be considered. Provided that the rotator cuff is competent and sufficient glenoid bone stock exists, we opt to resurface the glenoid in these patients.

HUMERAL HEAD OSTEONECROSIS

Although humeral head osteonecrosis is rare, it is the most common nontraumatic indication for shoulder hemiarthroplasty in our practice. In a large multicenter study, osteonecrosis was the underlying cause in only 5% of the primary shoulder arthroplasties performed.¹⁰ Many factors have been implicated as contributing to atraumatic osteonecrosis, including corticosteroid use, alcohol abuse, and hematologic disorders. Despite the influence of these factors, most cases of humeral head osteonecrosis are idiopathic.

Clinical Findings

Clinical findings in patients with atraumatic osteonecrosis range from solely subjective complaints of pain to glenohumeral crepitus and stiffness such as seen in primary osteoarthritis. Rotator cuff testing is usually normal but may be compromised by pain, especially in the later stages.

Imaging Findings

Radiographic classification of humeral head osteonecrosis has evolved from that described for the femoral head.^11,¹² Stage I is a preradiographic stage that requires diagnosis by magnetic resonance imaging or scintigraphy. Stage II is characterized by a zone of osteopenia surrounding a zone of relatively increased osseous density with the sphericity of the humeral head preserved (Fig. 6-8). Stage III is characterized by the presence of a subchondral fracture (the “crescent sign”); sphericity of the humeral head is preserved (Fig. 6-9). Stage IV corresponds to loss of sphericity of the humeral head as a result of collapse of the necrotic segment (Fig. 6-10). Stage V is characterized by loss of glenoid articular cartilage with secondary osteoarthritis (Fig. 6-11). Stage VI is characterized by osseous collapse of the humeral head with medialization of the humerus relative to the glenoid (Fig. 6-12).