Design: Systematic review and meta-analysis of randomised controlled trials (RCTs).
Data sources: Medline, Embase, Scientific Citation Index, CINAHL, Cochrane Library, and conference abstracts in the past 2 years from the British Society for Rheumatology, the European League Against Rheumatism, the American College of Rheumatology, and the Osteoarthritis Research Society International.
Subjects: 10 RCTs including 1712 patients with either symptomatic OA of the knee (6 trials) or hip/knee (3 trials) or multiple joints (1 trial).
Main outcome measures: (a) effect size (ES) for pain, stiffness, and functional scores from baseline to end point; (b) rate ratio (RR) and number needed to treat for clinical response rate and patient preference for treatment.
Results: Paracetamol was effective in relieving pain due to OA (ES = 0.21, 95% confidence interval (CI) 0.02 to 0.41). Non-steroidal anti-inflammatory drugs (NSAIDs) were better than paracetamol for pain relief (ES = 0.20, 95% CI 0.10 to 0.30). Clinical response rate was higher with NSAIDs than with paracetamol (RR = 1.24, 95% CI 1.08 to 1.41), and the number of patients who preferred NSAIDs was more than twice the number of those preferring paracetamol (RR = 2.46, 95% CI 1.51 to 4.12). NSAIDs were associated with more frequent gastrointestinal discomfort than paracetamol (RR = 1.35, 95% CI 1.05 to 1.75).
Conclusion: Paracetamol is an effective agent for pain relief due to OA. Although safer, it is less effective than NSAIDs. For safety reasons paracetamol should be the first line treatment, with NSAIDs reserved for those who do not respond.