A primary consideration in selecting the appropriate management in these patients is the etiology of the arthritis. The results of the treatment are variable and depend on whether the patient has a diagnosis of primary osteoarthritis (OA) or inflammatory arthritis.

An initial discussion with the patient about the arthritic process and its probable future implications is a critical first step in the treatment. This helps to ensure that the patients’ expectations are consistent with their prognosis. Education also promotes compliance with the treatment regimen and avoidance of detrimental activities. The goals of treatment and anticipated improvements should also be made clear. Many patients will not be able to regain previous levels of sporting and other recreational activities. If a patient is a manual laborer, the early realization that he or she may be unable to continue that previous occupation will facilitate early retraining.

Patients with arthritis must be taught to interpret their symptoms in order to provide information important in determining appropriate therapy. Developing a sense of control over their symptoms fosters a patient’s active participation in the treatment and helps improve the psychological component of the effects of the disease. An insightful essay by R. E. Jones41 discusses the importance of the personality and manner of the physician, as well as the quality of the physician-patient relationship in determining patient satisfaction. Applying the basic principles discussed by Dr. Jones will help physicians teach their patients how to manage and best enjoy the highest quality of life that is possible through current operative and nonoperative interventions for their arthritic shoulders.

Though in most patients suffering from glenohumeral arthritis the chief complaint is pain, the associated functional limitations are often a major concern. An impaired ability to dress and perform personal hygiene are particular problems. Occupational therapists can help these patients by providing assistive devices and by teaching alternative ways to accomplish these tasks. For patients who desire to continue their current employment, an analysis of specific job requirements is helpful in identifying and correcting problematic work practices.

Oral analgesics, such as salicylates, acetaminophen, and codeine, can be very effective in treating arthritic pain. The role of nonsteroidal anti-inflammatory drugs (NSAIDs) in treating patients with symptomatic inflammatory arthropathies is well established.⁴² However, for use in OA, it is controversial whether NSAIDs are any better than simple analgesia.²⁴ Nevertheless, a recent meta-analysis examining pain relief in patients with OA treated with NSAIDs versus acetaminophen, NSAIDs were statistically superior in providing relief of rest and walking pain.⁴⁷ This study also found an increased, though not statistically significant, rate of discontinuation secondary to adverse events in the NSAID group. Recently introduced COX-2 specific inhibitors have produced controversy, both with regard to efficacy as well as potential for exacerbating medical comorbidities. The side effects of these medications are not insignificant. They may have deleterious effects on articular cartilage in addition to the risk of untoward effects on gastric, renal, and liver function, which are particular problems in elderly patients.²⁴,²⁵,⁴² Recent work in the peer-reviewed literature has demonstrated a significant reduction in risk of ulcer development by administering a proton-pump inhibitor with NSAID or COX-2 inhibitor in older patients and those with an ulcer history.⁷⁶ Administration of opioids has recently been shown to result in significantly increased risks of cardiovascular events, fractures, and mortality than NSAIDs in elderly patients.⁸² These medications are not benign; physicians should seek to utilize any NSAID or other pain-relieving medication both judiciously and appropriately.

Nonoperative medical management of patients with intractable arthritic shoulder pain is difficult. A team approach may be necessary to provide adequate treatment of all factors affected by the disease. These specialists can manage depression, manipulate medication intake, inject trigger points, and perform nerve blocks. A neurologic workup may be necessary to rule out cervical radiculopathy, brachial plexopathy, reflex sympathetic dystrophy, or other disorders.

The diagnostic value of local anesthetic injections cannot be overstated and is a routine part of the initial patient workup. Appropriate steroid injections have also been documented to have significant therapeutic value.⁷,³⁴ Although the use of intra-articular steroids can be effective in abating symptoms in joints afflicted with RA, they are generally of comparatively limited value in OA, typically providing short-term relief of symptoms. Judicious use of these injections is warranted based on their potentially detrimental effects.⁵⁵

A prospective, randomized, blinded study by Dacre et al. evaluated the results of local steroid injections or physiotherapy in treating patients with painful or stiff rheumatoid shoulders.²² After controlling for age, gender, and disease severity, 60 consecutive patients were allocated into three groups to receive either local steroids, 6 weeks of physiotherapy, or both. Physiotherapy alone was just as effective as local steroid injections or a combination of these methods.

A recently published review of the peer-reviewed literature regarding the use of corticosteroid injections used for periarticular and intra-articular shoulder pain documented variable success.¹² Anecdotally, we have found that, if a single intraarticular water-soluble corticosteroid injection is helpful, then another can be performed at a 6-month interval (no more often than two injections per year). If the first injection provides minimal relief (assuming a truly intra-articular position of the injection—if there is any doubt, fluoroscopic imaging should be considered), then further corticosteroid injections are likely to have little potential benefit.

Intra-articular injections of hyaluronic acid preparations have been shown to be efficacious in the treatment of knee and hip OA. Recent studies have provided some evidence of possible chondroprotective and anti-inflammatory effects of hyaluronic acid.²,³⁶,³⁷,⁶³,⁹⁴ Another study demonstrated that intra-articular hyaluronic acid actually increased joint swelling, inflammation, and cartilage damage late in the chronic phase of arthritis.⁷² Viscosupplementation of the glenohumeral joint has been shown to improve pain and function.⁶,⁸,⁷³ There remains no peer-reviewed, double-blinded, prospective, randomized clinical trial documenting any benefit of such injections in patients with glenohumeral OA when compared to simple corticosteroid injections in the shoulder. Nevertheless, as with any other treatment modality, there may still be a benefit in certain patients who seek to avoid operative treatment for their glenohumeral arthritis. These indications have not yet been clarified.

Recently, many adjuvant nutritional supplements, or “nutraceuticals,” have been introduced for the amelioration of musculoskeletal complaints including arthritis. Animal work has demonstrated a reduction in the clinical, inflammatory, and histologic components of arthritis with the use of both glucosamine and chondroitin sulfates in combination. The apparent mechanism of glucosamine/chondroitin is to favorably shift cartilage biology toward matrix synthesis instead of degradation. Randomized controlled human studies have demonstrated conflicting results. A recent meta-analysis demonstrated no improvement in hip or knee arthritis symptoms and/or slowing of radiographic disease progression with glucosamine and/or chondroitin versus placebo.⁹³ No peer-reviewed report has documented the efficacy of these substances in patients with isolated glenohumeral arthritis, though patients with concomitant arthritis in other joints have anecdotally reported an overall relief. The current available evidence does not support routine use of these supplements in the treatment of OA.

Any patient with pain and loss of function attributable to glenohumeral arthritis is begun on a trial of anti-inflammatory agents, activity modification, and physical therapy after the education process described above. In our practice, corticosteroid injections are very rarely utilized, as we have found appropriate physical therapy and anti-inflammatory medications to be just as effective. If the patient continues to have pain after 6 weeks, we will change the anti-inflammatory medication and modify specific aspects of their physical therapy. If they develop progression of their arthritis characterized by further loss of motion and radiographic changes after completing this course of conservative therapy, nonoperative treatment is considered a failure and surgical management options are discussed.

Arthroscopic debridement or lavage for arthritis has been successfully used in the weight-bearing joints of the lower limb, particularly the knee.⁴⁰ In many cases, this may be the result of a strong placebo effect.⁵⁷ Most orthopedic surgeons know of anecdotal evidence suggesting that arthroscopic debridement for glenohumeral arthritis can benefit some patients; however, the indications and benefits are less clear. In some cases, arthroscopy will reveal previously unrecognized significant glenohumeral joint degeneration.²⁷ Cofield reported that, in eight patients with glenohumeral arthritis, the use of arthroscopy confirmed or modified the diagnosis, or altered the course of treatment, in all cases.¹⁶