To the Editor: We consider that it is important to comment on the views expressed by Langton et al on the limited value of paracetamol in the treatment of musculoskeletal pain.1

Langton et al recognise that paracetamol is widely recommended as first-line therapy to reduce chronic pain, but they largely dismiss its usefulness, noting that:

However, paracetamol is widely recommended as the first-line drug treatment in the management of osteoarthritis. This is based on its efficacy and safety as compared with nonsteroidal anti-inflammatory drugs (NSAIDs), including cyclooxygenase 2 (COX-2) inhibitors. This position is supported by published guidelines, including those of the American College of Rheumatology2 and the European League of Associations of Rheumatology (EULAR).3 In these guidelines, NSAIDs are recommended for use in moderate to severe osteoarthritis pain (American College of Rheumatology guidelines) or where the pain is unresponsive to paracetamol (EULAR guidelines). Our own Australian Therapeutic Guideline series and National Prescribing Service publications similarly recommend paracetamol as first-line treatment in osteoarthritis4 (see National Prescribing Service, Fact Sheet 8, October 2004 <www.nps.org.au>).

The efficacy of paracetamol in comparison with NSAIDs in patients with osteoarthritis has been demonstrated in patients treated for periods ranging from 3 weeks to 2 years, with total daily doses of paracetamol ranging from 2.6 g to 4.0 g,5 but this has been contentious.6,7 In a 2-year study involving 66 patients with osteoarthritis, Williams et al noted a higher withdrawal rate due to side effects in the naproxen group than in the paracetamol group, and slightly less efficacy in the paracetamol group.5 Pincus and colleagues reported that a third of patients receiving paracetamol continued on this treatment for more than 24 months, and that paracetamol was significantly less likely to be discontinued because of toxicity than NSAIDs.8 Thus, although paracetamol is on average less effective in pain reduction compared with NSAIDs,6 the difference in efficacy is small and a substantial proportion of patients can be treated satisfactorily and safely with paracetamol alone.9

Paracetamol remains the appropriate initial treatment for the management of osteo-arthritis. Other medications, such as NSAIDs or COX-2 inhibitors, can be added if patient response is unsatisfactory and the risk–benefit ratios are acceptable. When considering alternative options to rofecoxib, prescribers should also review the non-drug options, such as weight loss, physiotherapy, orthotics, and, where possible, opt for paracetamol first.10 Prescribers need to continue to be mindful of the potential for adverse effects with NSAIDs, particularly in high-risk patients or patients taking concomitant medications.

The serious public health problem of upper gastrointestinal tract bleeding caused by NSAIDs is a major consideration in the management of patients with osteoarthritis and becomes more of an issue in the elderly, many of whom have osteoarthritis.