Osteoarthritis (OA) is the most common form of arthritis in the United States. Prevalence in the United States was estimated at 27 million in 2008, an increase from 21 million in 1995.1 OA results from inflammatory and mechanical factors in susceptible individuals leading to joint pain, stiffness, and loss of range of motion.2 No therapies have been shown to alter the natural history of OA. In the absence of disease modifying osteoarthritis drugs (ie, DMOADs), treatment of OA is focused on controlling symptoms, especially pain. Non-pharmacologic therapies include weight loss, low-impact exercise, and physical therapy.2,3 Medical options include acetaminophen, nonsteroidal anti-inflammatory drugs (NSAIDs), tramadol, and opioids in selected patients.2,3 Topical therapies include capsaicin, methylsalicylate, and NSAIDs.2,3
Intra-articular (IA) injections provide an additional non-operative strategy for OA management when non-pharmacologic and medical therapies provide inadequate relief of symptoms. IA corticosteroids and hyaluronic acids (HA) are available for the treatment of OA, though the latter are approved by the US Food and Drug Administration solely for osteoarthritis of the knee.2 Both IA corticosteroids and HAs are recommended in guidelines for the management of knee OA.3,4 Joint replacement surgery should be reserved for those patients who fail pharmacologic and non-pharmacologic modalities, including an adequate trial of injection therapy.
The efficacy of IA hyaluronates for the treatment of symptomatic knee OA has been previously reviewed.2,5 Similar to the findings for IA corticosteroids for knee OA, a Cochrane analysis of viscosupplementation for the treatment of knee OA found HA derivatives as a class to be effective.5,6 There is increasing interest in the use of HAs for OA at sites other than the knee. Though preliminary reports suggest efficacy for shoulder and hip OA, these indications are currently considered off label.7,8
Several questions remain regarding the use of HAs for OA. Should the initial injection for OA be a corticosteroid or an HA? Is there evidence to support use of a particular HA product over its competitors? Can injection therapies be combined for OA? This column will explore these concerns based on the currently available evidence and provide comments based on the author’s practice.
SELECTION OF INITIAL INTRA-ARTICULAR AGENT FOR KNEE OSTEOARTHRITIS
Corticosteroids and HAs are commonly used for the management of knee OA not responding to more conservative therapy. Few direct comparisons of IA corticosteroids to HA injections have been performed. Therefore, current guidelines make no recommendations as to which class should be initially employed once the decision is made to use injection therapy for knee OA.
In one of the largest comparison studies reported, Leopold and colleagues9 randomized 100 patients with knee OA to receive a 3 injection series of Hylan G-F 20 or a corticosteroid injection. Both groups were followed for a total of 6 months and the corticosteroid group was allowed a second corticosteroid injection at any time during the study period. At study conclusion, there were no significant differences between the groups as assessed by 3 different pain scales commonly used to measure response in knee OA trials.
The Osteoarthritis Research Society International (OARSI) has examined treatment effects of both corticosteroids and HAs for knee OA in a review of research published through January 2009.4 The effect size of corticosteroids on knee OA was estimated at 0.58, compared with 0.60 for HA derivatives (0.5 indicates a moderate effect). The number needed to treat (NNT) for each modality was also similar: 5 for corticosteroids and 7 for HA injections. This analysis suggests IA corticosteroids and HA can be expected to deliver similar results in clinical practice.
Safety is an important consideration in choosing an initial agent for injection. In general, the rate of post-arthrocentesis septic arthritis is low and has been estimated at 1 in 14,000 to 1 in 50,000 following corticosteroid injections.10 Albert and colleagues11 described 2 cases of septic knee arthritis following intra-articular hyaluronate injection in elderly OA patients, but no estimates are currently available specifically for the rate of post-hyaluronate injection septic arthritis. Patients should be informed of the risk of infection when counseled about this procedure, because infection can occur following injection with either corticosteroid or HA.
In addition to septic arthritis, risks of HA injection include pseudoseptic injection reaction and flare of crystalline arthritis. Attacks of gout and calcium pyrophosphate dehydrate arthritis (ie, pseudogout) have been described following HA injection.12,13 Noninfectious inflammatory reactions have also been described and are more common with hylan G-F 20 injections.14 When HA injection results in acute joint inflammation, arthrocentesis with synovial fluid analysis is mandatory. Synovial fluid must be sent for Gram stain, culture, and cell count. Crystal analysis should also be performed to exclude gout or pseudogout. Empiric treatment for septic arthritis may be considered until results of these studies are available. Both crystalline arthritis and the pseudoseptic inflammatory reaction respond favorably to a corticosteroid injection once infection has been excluded.