This study is a prospective, randomized, double-blind study to compare the efficacy and safety of 10 mg/kg infliximab with those of 3 mg/kg infliximab treatment in methotrexate-refractory rheumatoid arthritis patients. After the patients received 3 mg/kg infliximab infusion at weeks 0, 2, and 6, they were randomly assigned to be administered 3, 6 or 10 mg/kg infliximab every 8 weeks from week 14 to 46. Mean American College of Rheumatology improvement (ACR-N) at week 54, the primary endpoint, was 51.3% and 58.3% for the 3 mg/kg and 10 mg/kg groups, respectively, with a statistically significant difference. Treatment with 10 mg/kg was found to be remarkably beneficial in patients who had not responded to three infusions with 3 mg/kg at week 10. The median changes in the modified Sharp score were 0.0 in the two groups. There were no significant differences in the incidences of adverse events between the groups. In patients who achieved better clinical response or greater inhibition of progression of joint damage, trough serum infliximab level was significantly higher than in patients who did not. The magnitudes of both efficacies were correlated with the trough serum infliximab level (ClinicalTrials.gov number: NCT00691028).