Background: Topical and oral nonsteroidal anti-inflammatory drugs (NSAIDs) provide short-term relief of chronic knee pain. Topical NSAIDs are believed to be as effective as oral agents, but with fewer adverse effects. Underwood and colleagues compared the outcomes of advice to older patients about using oral or topical ibuprofen.

The Study: Study participants were 50 years or older. To be eligible, participants must have consulted a general practitioner in one of the 26 participating practices in England. Participants must have had osteoarthritis, or knee or leg pain in the preceding five years, or have been prescribed oral or topical NSAIDs during the previous year. Participants were required to have had troublesome knee pain on most days for at least one month, as well as knee pain for more than three months during the previous year. Patients who had knee replacements were excluded from the study.

Participants were asked to refrain from using any oral or topical NSAIDs for one week before study intake. Patients were clinically assessed based on the American College of Rheumatologists' clinical criteria for knee osteoarthritis. Baseline clinical and demographic data were gathered on entry to the study.Participants were offered the choice of joining a randomized controlled trial of oral versus topical NSAIDs, or a preference study in which they selected their own form of NSAID therapy. Postal questionnaires were sent to participants at three, six, 12, and 24 months. Research nurses conducted clinical assessments and blood tests at 12 and 24 months. The primary outcome measure was the Western Ontario and McMaster Universities (WOMAC) score, which is based on knee pain, stiffness, disability, and overall functioning. Secondary outcomes included measures of chronic pain and related disability.

Results: The randomized trial included 282 patients, and 303 patients entered the preference study. The average age was 64 years. Participants were comparable across study groups, but those in the topical treatment group of the preference study tended to be older and of lower socioeconomic class than other participants.

Follow-up was done in 83 percent of participants at 12 months, and 55 percent at 24 months. Changes in the WOMAC scores at 12 months were small, but equivalent across all groups. The only finding that achieved statistical significance was a borderline improvement in WOMAC score at 24 months for oral medication over topical medication in the randomized trial. The oral group in the randomized trial experienced slightly less pain-related disability at three months, but this was not sustained to 12 months.

The groups did not differ significantly in reported major or minor adverse effects, but respiratory adverse effects were less common in the topically-treated groups. A statistically significant difference in favor of topical therapy was recorded in creatinine concentrations in the randomized trial.

After 12 months, 11 percent more participants in the topical group in the randomized trial changed treatments because of inadequate pain relief. Five percent of those randomized to oral treatment obtained prescriptions for topical agents. However, 37 percent of those randomized to topical treatment and 26 percent of those selecting topical treatment obtained prescriptions for oral NSAIDs.The number of participants receiving rescue medication was greater among those who selected oral therapy in the preference study than those who chose the topical therapy. This difference almost achieved statistical significance.

Conclusion: Advising patients to use topical NSAIDs for osteoarthritis of the knee was associated with similar clinical outcomes but slightly fewer adverse effects than advising patients to use oral versions of the same medications. The authors conclude that topical NSAIDs may be a useful alternative to oral NSAIDs for symptoms of osteoarthritis of the knee.