Patients with osteoarthritis of the hand may benefit from assistive devices and instruction on techniques for joint protection; splinting is beneficial for those with symptomatic osteoarthritis of the first carpometacarpal joint. Patients with mild to moderate osteoarthritis of the knee or hip should participate in a regular exercise program (eg, a supervised walking program, hydrotherapy classes) and, if overweight, should lose weight. The use of assistive devices (eg, a cane on the contralateral side) can improve functional status.
1. Oral nonsteroidal anti-inflammatory drugs (NSAIDs)
NSAIDs (see Table 5–5) are more effective than acetaminophen for osteoarthritis but have greater toxicity. NSAIDs inhibit cyclooxygenase (COX), the enzyme that converts arachidonic acid to prostaglandins. Prostaglandins play important roles in promoting inflammation, but they also help maintain homeostasis in several organs—especially the stomach, where prostaglandin E serves as a local hormone responsible for gastric mucosal cytoprotection. COX exists in two isomers—COX-1, which is expressed continuously in many cells and is responsible for the homeostatic effects of prostaglandins, and COX-2, which is induced by cytokines and expressed in inflammatory tissues. Most NSAIDs inhibit both isomers. Celecoxib is the only selective COX-2 inhibitor currently available in the United States.
Gastrointestinal toxicity, such as gastric ulceration, perforation, and gastrointestinal hemorrhage, are the most common serious side effects of NSAIDs. NSAIDs can also affect the lower intestinal tract, causing perforation or aggravating inflammatory bowel disease. The overall rate of bleeding with NSAID use in the general population is low (1:6000 users or less) but is increased by the risk factors of long-term use, higher NSAID dose, concomitant corticosteroids or anticoagulants, the presence of rheumatoid arthritis, history of peptic ulcer disease or alcoholism, and age over 70. Proton pump inhibitors and histamine type-2 receptor antagonists reduce the incidence of serious gastrointestinal toxicity and should be used for patients with risk factors for NSAID-induced gastrointestinal toxicity. Patients who have recently recovered from an NSAID-induced bleeding gastric ulcer appear to be at high risk for rebleeding (about 5% in 6 months) when an NSAID is reintroduced, even if prophylactic measures (such as proton pump inhibitors) are used. Compared with nonselective NSAIDs, celecoxib is less likely to cause upper gastrointestinal tract adverse events, including bleeding.
Acute liver injury from NSAIDs is rare, occurring in about 1 of every 25,000 patients using these agents.
All of the NSAIDs, including aspirin and celecoxib, can produce renal toxicity, including interstitial nephritis, nephrotic syndrome, prerenal azotemia, and aggravation of hypertension. Hyperkalemia due to hyporeninemic hypoaldosteronism is seen rarely. The risk of renal toxicity is low but is increased by the following risk factors: age older than 60 years, history of kidney disease, heart failure, ascites, and diuretic use.
All NSAIDs, except the nonacetylated salicylates and celecoxib, interfere with platelet function and prolong bleeding time. Aspirin irreversibly inhibits platelet function, so the bleeding time effect resolves only as new platelets are made. In contrast, the effect of nonselective NSAIDs on platelet function is reversible and resolves as the drug is cleared. Concomitant administration of a nonselective NSAID can interfere with the ability of aspirin to acetylate platelets and thus may interfere with the cardioprotective effects of low-dose aspirin. The US Food and Drug Administration (FDA) has warned that all NSAIDs can increase the risk of myocardial infarction and stroke in patients with or without risk factors for heart disease or known heart disease. While the cardiovascular risk is related to the dose and duration of treatment, stroke and myocardial infarction can occur within the first week of treatment. Cardiovascular risks associated with naproxen, ibuprofen, and moderate dose celecoxib (200 mg orally daily) are comparable.
Chondroitin sulfate and glucosamine, alone or in combination, are no better than placebo in reducing pain in patients with knee or hip osteoarthritis.
Topical NSAIDs (eg, 4 g of diclofenac gel 1% applied to the affected joint four times daily) appear more effective than placebo for knee and hand osteoarthritis and have lower rates of systemic side effects than with oral NSAIDs. Topical NSAIDs are preferred for patients 75 years of age and older. Topical capsaicin may be of benefit for osteoarthritis of the hand or the knee.
Acetaminophen is not recommended given that its impact on pain is frequently negligible and hepatotoxicity can occur from high doses. Opioids are generally not appropriate for the long-term management of pain due to osteoarthritis.
4. Intra-articular injections
Many patients with moderately severe osteoarthritis of the knee who do not respond to NSAIDs receive intra-articular injections of corticosteroids, hyaluronate, or platelet-rich plasma. Although each of these can temporarily reduce pain, none has convincingly produced long-term benefits in reducing pain or preserving function. A 2-year controlled trial demonstrated that injecting the knee with triamcinolone every 6 months was no more effective than injecting saline in reducing knee pain. The American College of Rheumatology does not recommend corticosteroid injections for osteoarthritis of the hand.
For patients with osteoarthritis in multiple joints who either have not responded to or cannot use NSAIDs, the selective serotonin and norepinephrine reuptake inhibitor duloxetine, 30–60 mg orally daily, can reduce pain. Nausea occurs in 6–15% of patients.
Total hip and knee replacements provide excellent symptomatic and functional improvement when involvement of that joint severely restricts walking or causes pain at rest, particularly at night. Arthroscopic surgery for knee osteoarthritis is ineffective.