Background Although methotrexate is recommended as the anchor and initial drug for treatment of RA, a considerable number of patients present major side effects or have inefficacy to this drug. Hence, it is necessary to assess other synthetic DMARD's which might be used as initial therapy with at least the same efficacy and less side effects.

Objectives Evaluate the efficacy and side effects of methotrexate or leflunomide as initial therapy in RA in a meta-analysis and systematic review of published randomized controlled trials (RCT).

Methods A systematic search of publications assessing efficacy and safety in RCT comparing methotrexate, leflunomide or placebo in the Cochrane Library, MEDLINE, EMBASE, PUBMED and Science Citation Index were performed. A total of 1,691 papers were found, after a review of their abstracts, 73 articles met inclusion criteria for whole review. In a structured workshop, authors verified that each trial clearly define variables useful for meta-analysis, mainly number and demographic characteristics of participants, previous treatments, interventions, follow-up, assignment of treatment; as well as outcome measures, such as ACR index change, changes in HAQ, serum C-reactive protein (CRP), and side effects including deaths. After this approach, 52 trials were excluded, mainly due to lack of information, or because one study branch included a biologic agent. In a secondary revision round, 21 RTC were assessed and only 8 could be included for the final analysis; reasons for exclusion of 13 trials were short follow-up, unclear clinical, or different radiologic, molecule expression or biopsy outcomes, or DMARDs combinations with one of methotrexate or leflunomide. These 8 trials had adequate Jadad scores.

Analysis Treatment effects were analysed with the RevMan Analyses V5 software; for discrete data we calculate odds ratios, and for continuous data their means were compared. Forest graphics were assess the heterogeneity effects among RCT with χ² and I² statistics. Significant heterogeneity was defined with p>0.10 (χ²) or I²<25%. Assessment of bias was made according to The Cochrane Handbook for Systematic Reviews of Interventions.

Results The finally analysed RCTs included a total of 1,990 patients. In the comparison of ACR20 achievement, leflunomide and methotrexate, as it is expected, showed greater efficacy than placebo, (OR 3.96; IC 95% 3.12-5.01, and OR 3.44; IC 95% 2.72-4.34, respectively). Moreover, leflunomide was superior to methotrexate at 52 weeks, OR 1.21; IC 95% (1.01-1.45; I² 2%). On the other hand, leflunomide trend to show less side effects, mainly of infectious and gastrointestinal type, OR 1.0; IC 95% (0.74-1.35). Assessment of change in HAQ-Di score and CRP could not be performed because only two studies had these measurements.

• Download figure
• Open in new tab
• Download powerpoint

Conclusions Leflunomide seems to be more efficacious than methotrexate as initial treatment of RA, gastrointestinal adverse effects have lower presentation with this drug.

References

1. Smolen JS, et al. EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease modifying antirheumatic drugs. Ann Rheum Dis 2010; 69:964–975.

2. Julian P. T. Higgins, Sally Green. Cochrane Handbook for Systematic Reviews of Interventions. Chocrane Handbook Series, Wiley BlackWell.

Disclosure of Interest None declared