Article Text


FRI0039 An assessment of leflunomide for patients with rheumatoid arthritis treated in routine clinical practice
  1. JA Teir,
  2. AD Woolf,
  3. MJ Davis
  1. Rheumatology, Royal Cornwall Hospital, Truro, Cornwall, England, UK


Background Phase 3 clinical trails used for licensing of new agents are not always reflective of ‘day to day’ clinical practice.

Objectives We have examined the use and effect of Leflunomide in the routine clinic situation.

Methods From December 1999 until May 2000 20 patients with rheumatoid arthritis commenced on Leflunomide had a full clinical assessment in a general rheumatology clinic. All had joint scores (28 swelling and tenderness), health assessment questionnaire (HAQ), VAS pain, patient and physician global and C-reactive protein (CRP) measured at initiation of treatment and follow up visit. joint scores were performed by clinic nurses,5 in total, given basic training in the technique. the mean follow up was 6 months (2–9 months) with the above the American College of rheumatology (ACR) criteria for response was calculated.

Results At final assessment the mean SJC had reduced from 14.5 to 11.0 (p < 0.02), mean TJC 14.1 to 10.0 (p < 0.02), mean CRP 55.3 to 30 (p < 0.01) and mean HAQ from 2.3 to 2.2(p < 0.9). Only 5(25%) however achieved an ACR20 of which 2 were ACR50. At mean follow up of 6 months 7(35%) had discontinued treatment because of adverse reaction. 6 were as a result of bowel disturbance and 1 because of headache. in this small group no patients had serious abnormalities in either their liver function or full blood counts. 13(65%) were continuing with therapy content with its efficacy.

Conclusion The ACR 20 response is clearly less than the published trails (52%) yet the perceived patient efficacy is similar. The difference related to the difficulty in performing, reliably, the joint scores in routine clinical practice. this ought to considered when suggesting these measures should be performed outside of the research setting. The incidence of bowel upset is higher than the phase 3 study (13%). This may be attributable to different patient populations with fewer exclusion criteria in clinical practice.


  1. Smolen JS, et al. Lancet 1999;353:259–66

Statistics from

Request permissions

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.