Commentary: controversies in NICE guidance on osteoarthritis

BMJ 2008; 336 doi: (Published 28 February 2008) Cite this as: BMJ 2008;336:504
  1. Nicola Dalbeth, rheumatologist and senior lecturer1,
  2. Bruce Arroll, professor and head of department2
  1. 1Department of Medicine, University of Auckland, Private Bag 92019, Auckland, New Zealand
  2. 2Department of General Practice and Primary Health Care, University of Auckland
  1. Correspondence to: Bruce Arroll b.arroll{at}

The major focus of the guideline is management of established pain and disability in patients with existing osteoarthritis. It raises several conundrums for clinicians. Firstly, it identifies as core treatments those that are low risk and low cost. However, these treatments are also generally low in efficacy with low or negligible effect sizes.1 Of the core treatments, only exercise therapy has a moderate effect size for treating pain in osteoarthritis, which is similar to the effect size of the more risky and costly oral non-steroidal anti-inflammatory drugs (NSAIDs). Although these core treatments should be integrated into osteoarthritis management, in clinical practice they are rarely sufficient as sole treatments. Furthermore, many patients with osteoarthritis do not adhere to long term exercise programmes.2 Analysis of patient acceptability of the guideline’s core interventions would be of great interest.

Secondly, the inclusion of topical NSAIDs as a preferable pharmacological treatment for hand and knee osteoarthritis should be emphasised. This treatment is often underused in osteoarthritis management.3 However, these agents are supported by an excellent safety profile and data supporting clinical efficacy, and are rightly recommended in the guideline.

Thirdly, the guideline does not recommend glucosamine or chondroitin supplements for osteoarthritis pain. Although the overall evidence for glucosamine is variable, some high quality studies have shown that glucosamine sulphate 1500 mg daily has benefit for knee osteoarthritis pain, and this agent has an excellent safety profile.4 Furthermore, we have been impressed at the willingness of elderly patients to pay for this agent in a jurisdiction (New Zealand) where patients will traditionally pay only for medications that are symptomatically effective.

Fourthly, oral NSAIDs are included as adjunctive pharmacological treatment when paracetamol and topical NSAIDs are insufficient. Protection of the gastrointestinal tract with a proton pump inhibitor is recommended for all patients taking NSAIDs, regardless of age, based on two recently published randomised controlled trials.5 Importantly, these studies recruited only patients at high risk of ulcers (≥60 years with or without a history of ulcers) who were using daily NSAIDs. Most patients requiring NSAIDs for osteoarthritis are likely to be aged over 60 years and require daily NSAIDs. However, whether younger low risk patients requiring occasional NSAIDs should be routinely coprescribed a proton pump inhibitor is currently unclear.

Fifthly, the guideline tackles the question of referral for joint replacement, emphasising the importance of surgical referral before development of prolonged disability and severe pain. The guideline states that patient factors such as age, smoking, obesity, and comorbidities should not be a barrier to referral for surgery. Whether referral of very old patients and those with serious medical comorbidity (including obesity) will lead to higher rates of surgery in these groups, or indeed greater perioperative complication rates, remains uncertain.

Finally, this guideline has important resource implications, emphasising the need for adequate funding and availability of community exercise facilities, educators, physiotherapists, and dietitians for patients with osteoarthritis in order to deliver core treatments. In addition, adequate access to hand therapists, occupational therapists, podiatrists, orthotists, and orthopaedic surgeons is needed, to ensure that recommended adjunctive therapies are available to patients with osteoarthritis when indicated.


  • doi: 10.1136/bmj.39490.608009.AD
  • Contributors: Both authors did a literature review and wrote the article. BA is the guarantor.

  • Competing interests: None declared.


View Abstract

Log in

Log in through your institution


* For online subscription