Statistics from Altmetric.com
Guidelines for the treatment of osteoarthritis (OA) have several purposes, but a major one relates to its use by regulatory or medical insurance agencies that use it to help guide their response to requests for drugs to treat patients.
Data have recently been published which record the efficacy of non-steroidal anti-inflammatory drugs (NSAIDs) over paracetamol (acetaminophen) and report patients' preferences for NSAIDs rather than paracetamol in the treatment of OA.1–7 Furthermore, inflammation is a common accompaniment of moderate to severe OA.8–10 Therefore it would seem appropriate to consider the use of NSAIDs as the primary treatment in the group of patients with OA with moderate to severe pain or inflammation as part of their disease.
The EULAR Guidelines state that paracetamol is the first line of treatment for all patients with OA,11 despite the recent evidence that NSAIDs are more efficacious than paracetamol and that paracetamol is associated with more gastrointestinal toxicity than was previously thought.12
Would it not be more appropriate to suggest that patients who have moderate to severe pain and/or OA with inflammatory components be given NSAIDs as the first line of treatment, leaving paracetamol for symptomatic use in those patients with lesser degrees of pain or lack of overt inflammation?
The pro's and con's of oral non-steroidal anti-inflammatory drugs (NSAIDs) and paracetamol are certainly of interest and relevance to the management of osteoarthritis (OA). However, the following points are relevant to the suggestions contained in Dr Furst's letter:
The EULAR recommendations reflect an evidence based approach to key clinical questions. They clearly specify the end of 1998 as the time up until which published evidence was included for review. The recommendations will be revised in the light of more recent publications in the future and will then include the studies referenced by Dr Furst. It is clearly impractical regularly to change individual recommendations in response to every new publication—certain time points need to be taken to “close the door” on data while the formal review process takes place.
Again it is clearly stated in the EULAR document that the package of care offered to a patient with knee OA must take into account not only efficacy and side effects of individual interventions but also patient attitudes and preferences, comorbidity, concurrent drug treatment, costs, and availability. It will indeed be interesting to see if the EULAR Task Force do change their recommendations in the light of these recent data in the way suggested by Dr Furst. For example:
In the recent study by Pincus it is noteworthy that although 57% of patients thought that Arthrotec was better than paracetamol, 20% thought that paracetamol was better and 22% reported no difference.1 In other words almost half the patients found no perceived benefit from Arthrotec. Side effects were significantly more common with Arthrotec. Furthermore, this was a short term study of six weeks. The few long term studies show that many patients stop their NSAID—in one study only 15–20% of patients with OA continued their NSAID for one year.2
Traditional oral NSAIDs indeed kill a significant number of patients each year from gastrointestinal bleeding/perforation. Over half of the patients with symptomatic OA are in the high risk category for NSAID associated mortality (elderly, with comorbidity). Although there are now data that paracetamol may have a gastrointestinal risk, it is less than that caused by traditional NSAIDs. There are no renal or cardiovascular side effects from paracetamol, but there are from NSAIDs.
NSAIDs remain expensive compared with paracetamol.
Data show that NSAIDs (at least indometacin) may hasten cartilage and bone attrition3—there are no data that paracetamol is such a negative “DMOAD”.
It is for reasons such as this that most doctors keep oral NSAIDs in reserve for patients in whom the safer and cheaper drug, paracetamol, has failed.4
It is perhaps unfortunate that many of us get hung up on the order of drug selection. In practice, of course, most patients receive a package of care, not a single treatment. Many patients take both NSAIDs and paracetamol (often as self medication of over the counter drugs)—hopefully in addition to exercise, weight reduction, pacing of activities, etc, etc. The argument about “first use” is therefore irrelevant to most patients. Unfortunately, few studies use a factorial design to better inform us of effect sizes of combined treatments. Few also examine predictors of response. These points are again made in the EULAR document.
We would take issue with the suggestion that a major purpose of “guidelines” is that they be used by regulatory or insurance agencies. This may be true in America but is not a rationale for producing recommendations in Europe. The objectives of the EULAR recommendations are clearly listed. They summarise evidence to help directly (not indirectly through agencies) guide decision making by practitioners and patients. They also highlight gaps in our knowledge and inform the future research agenda.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.