Paul Creamer

Pain is the most important symptom of osteoarthritis (OA) and the reason why individuals seek medical treatment. The anatomic cause is unclear and is likely to vary between individuals. Recent work confirms the heterogeneity of pain in OA with differences in severity, location, precipitating and relieving factors, and response to intra-articular anesthetic. Nonpharmacologic treatment of OA is important and evidence is now accumulating for interventions such as aerobic exercise, quadriceps exercises, footwear modification, education, and social support. Analgesia remains the first choice drug therapy: compounds more potent than acetaminophen are now available and effective. New cyclooxygenase-2 (COX-2) inhibitors may have a role in subjects for whom simple analgesia is inadequate. Glucosamine is a simple, safe product that appears to have a weak pain-relieving effect, and intra-articular hyaluronate injections may also have a limited role. Recent community studies confirm the benefit of joint replacement in OA, though a number of questions remain about the timing, indications, and alternatives to surgery.

Acupuncture has been found to be beneficial in the treatment of patients with knee osteoarthritis. However, response among such patients is highly variable. Identification of subjects with greater response would facilitate a more rational use of acupuncture. To examine the relationship between demographic and psychosocial variables and response to acupuncture as defined by reduction in pain and disability at the end of an 8-week course of treatment. Retrospective study. Outpatients attending rheumatologists or primary care physicians. 37 patients with symptomatic knee osteoarthritis who had previously participated in a controlled trial using acupuncture were recalled for an interview approximately 1 year later. Structured interview, questionnaire completion, and an examination. Depression, anxiety, helplessness, self-efficacy, and fatigue were measured by standard instruments. Knee examination and assessment of pain threshold were measured by dolorimetry. Response at 8 weeks was significantly related to duration of symptoms. A statistically nonsignificant trend was found for older and more educated subjects to have a better response; anxiety and fatigue were found to be inversely related to response (also statistically nonsignificant). Subjects with localized medial pain had significantly better response in terms of pain and disability than did subjects with generalized knee pain. Other than a weak relationship with anxiety (at 8 weeks only), no evidence of a link between psychosocial variables and response to acupuncture was found. Prospective studies are needed to confirm these results.