Dr Gillian Hosie, past President of the Primary Care Rheumatology Society and a GP in Glasgow


It has been estimated that around 2000 deaths each year are caused by the side-effects – mainly gastrointestinal – of non-steroidal anti-inflammatory drugs (NSAIDs).1

Many GPs aim to offset the potential problems caused by NSAIDs by limiting their prescribing or co-prescribing gastroprotective agents (GPAs).

In its recent guidance,2 NICE has recognised, in accordance with modern rheumatology thinking, that neither cyclo-oxygenase (Cox II) selective inhibitors nor standard NSAIDs should be used in the long term unless clinically necessary, and that repeat prescriptions of these drugs should be monitored and the drugs discontinued as soon as is feasible.

There are many ways of managing osteoarthritis (OA) without using NSAIDs. Non-drug therapies and simple analgesics should be the mainstay of therapy in OA.

In rheumatoid arthritis (RA), it is hoped that increasingly early intervention with disease-modifying anti-rheumatic drugs (DMARDs) will decrease the long-term use of NSAIDs. If a good response can be obtained to DMARD therapy, NSAID prescription can be reduced or even withdrawn. Patients can then substitute simple analgesics for NSAIDs, keeping anti-inflammatory drugs for flares of their condition.3

NICE has also tried to clarify the place of GPAs in the prescription of both Cox II selective inhibitors and standard NSAIDs.

The guidance states that not only is there no evidence to justify the prescription of GPAs with Cox II inhibitors, but also that co-prescription of GPAs has been shown to be only partially effective in the prevention and treatment of NSAID gastrointestinal side-effects.

Another area of concern to many doctors is the use of Cox II inhibitors in patients with cardiovascular disease. The VIGOR trial4 suggested an increase in myocardial infarction (MI) in patients on rofecoxib compared with those on naproxen. (These study patients were not allowed low-dose aspirin.) In the CLASS study5 of celecoxib vs ibuprofen or diclofenac, patients were allowed low-dose aspirin. In these patients the concomitant use of aspirin reduced the gastric protection thought to be afforded by celecoxib to non-significance.

NICE has suggested that doctors should exercise caution in prescribing Cox II selective inhibitors for patients with pre-existing cardiovascular disease. If such a patient is thought to require an anti-inflammatory drug, should we prescribe low-dose aspirin with a Cox II inhibitor, thus reducing the gastroprotective effect, or do we omit aspirin, potentially increasing the risk of MI, or prescribe a standard NSAID with a GPA?

The result is that often the GP has to balance the risks and benefits of prescribing certain drugs in certain circumstances.

Until this recent guidance fom NICE, many authorities were advising GPs not to prescribe Cox II selective inhibitors and these drugs were excluded from many formularies. The fact that NICE has recommended their use in certain well defined categories will make it much easier for GPs to tailor prescription of these drugs to appropriate patients.

Drug costs may rise with increased prescription of Cox II selective inhibitors and this will obviously impact on prescribing budgets. The increased costs, however, may be offset by a reduction in the costs of GPAs and a reduction in the number of prescriptions for anti-inflammatory agents by more appropriate use of non-drug therapies and simple analgesics, particularly for OA.


  1. Blower A et al. Aliment Pharmacol Ther 1997; 11(2): 283-91.
  2. Guidance on the use of cyclo-oxygenase (Cox) II selective inhibitors, celecoxib, rofecoxib, meloxicam, and etodolac for osteoarthritis and rheumatoid arthritis. Technology Appraisal Guidance – No 27. NICE, July 2001.
  3. Management of Early Rheumatoid Arthritis. SIGN No 48. Edinburgh: SIGN, 2000.
  4. Bombardier C et al Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis: VIGOR Study Group. N Engl J Med 2000; 342(21): 1520-8.
  5. Silverstein FE et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis. The CLASS Study, a randomised controlled trial. JAMA 2000; 284(10): 1247-55.

Guidelines in Practice, September 2001, Volume 4(9)
© 2001 MGP Ltd
further information | subscribe