Objective: To determine whether demographic, clinical, and laboratory variables measurable at baseline predict the clinical efficacy or major toxic effects of methotrexate (MTX) therapy in children with chronic arthritis.
Study design: Patient eligibility criteria: (1) monitored in our unit between 1986 and 1996 with a diagnosis of chronic arthritis and (2) treatment with MTX as the sole second-line agent and for at least 6 months. Outcomes investigated: (1) short-term (6-month) clinical response, (2) complete disease control, (3) disease relapse after MTX discontinuation after complete disease control, (4) aminotransferase elevation, (5) gastrointestinal toxicity. Independent variables that showed significant results with univariate tests or were clinically relevant for each outcome underwent multiple logistic or Poisson regression analyses.
Results: Eighty patients were available for analysis. The disease onset subtype was systemic in 37 patients, polyarticular in 20 patients, and oligoarticular in 23 patients (all with polyarticular course: extended oligoarticular subtype). The extended oligoarticular subtype was the best predictor for both the short-term clinical response (odds ratio 6.80, P =.02) and, together with a better functional ability, the complete disease control (rate ratio 3.85, P =.03 and rate ratio 3.29, P =.006, respectively). Patients with this subtype of chronic arthritis tended to have earlier, and more frequently, a disease relapse after MTX discontinuation. Thrombocytosis was the only significant risk factor for liver biochemical abnormalities (rate ratio 2.94, P =.008), whereas no variable yielded significant results for gastrointestinal toxicity.
Conclusion: Patients with extended oligoarticular chronic arthritis were more likely to benefit from MTX therapy and to have a relapse after treatment discontinuation, suggesting that MTX is distinctly more effective in this subset of chronic arthritis.