STUDY OBJECTIVE: To determine whether methotrexate has anti-inflammatory activity in refractory inflammatory bowel disease. DESIGN: Nonrandomized, open-label, preliminary trial of methotrexate along with standard medications for 12 weeks. SETTING: Referral-based gastroenterology practice. PATIENTS: Twenty-one patients with refractory inflammatory bowel disease (14, Crohn disease; 7, chronic ulcerative colitis); 17 taking variable doses of corticosteroids and 14 on sulfasalazine or metronidazole. Of the 21 patients, 10 had previously failed azathioprine or 6-mercaptopurine trials. INTERVENTIONS: Sulfasalazine and metronidazole were continued and prednisone dose was tapered according to clinical response. Methotrexate was given as a 25-mg intramuscular injection weekly for 12 weeks, then switched to a tapering oral dose if a clinical and objective improvement was noted. MEASUREMENTS AND MAIN RESULTS: Sixteen of twenty-one patients (11 of 14 patients with Crohn disease, 5 of 7 patients with chronic ulcerative colitis) had an objective response as measured by disease activity indices (modified Crohn's Disease Activity Index, 13.3 to 5.4 [P = 0.0001], Ulcerative Colitis Activity Index, 13.3 to 6.3 [P = 0.007]). Prednisone dosage decreased from 21.4 mg +/- 5.6 (SEM) to 5.5 mg +/- 2.0; P = 0.006 and 38.6 mg +/- 6.35 to 12.9 mg +/- 3.4; P = 0.01, respectively. Five patients with Crohn colitis had colonoscopic healing and 4 had normal histology at 12 weeks. In contrast, none of the 7 patients with ulcerative colitis had normal flexible sigmoidoscopies, despite histologic improvement in 5. Side effects included mild rises in transaminase levels in 2 patients, transient leukopenia in 1, self-limited diarrhea and nausea in 2 patients, and 1 case each of brittle nails and atypical pneumonitis. CONCLUSIONS: Although this pilot study is encouraging, further work is needed before methotrexate can be recommended for inflammatory bowel disease.