Aims Enteric bacteria are thought to contribute to the pathogenesis of Crohn's disease, and antibiotics may be an effective therapy. This study examines the efficacy of the nonsystemic (<0.4% absorbed) antibiotic rifaximin for inducing remission in patients with Crohn's disease. Methods Data from charts of patients with Crohn's disease who received rifaximin between 2001 and 2005 and had a Crohn's disease activity index score >= 220 at the time of rifaximin initiation were analyzed. The use of concomitant medications (e.g., steroids, anti-inflammatory agents) was allowed. Results In the 68 patient charts analyzed, the median duration of rifaximin treatment was 16.6 weeks, and the majority of patients (94%) received rifaximin 600 mg/day. Eighteen patients (26%) received rifaximin monotherapy, and 31 patients (46%) received concomitant steroids. The median baseline Crohn's disease activity index score at the time of rifaximin initiation was 265 (range, 220-460), and the mean duration of Crohn's disease was 17 years (range, 1-50 years). Crohn's disease remission occurred in 65% of patients. A 70% remission rate was achieved in patients who did not receive steroids, versus 58% in patients who received steroids. Clinical improvements continued 4 months after rifaximin initiation. Remission was achieved in 67% of patients who received rifaximin monotherapy. Conclusions Rifaximin therapy was associated with clinical improvement in patients with Crohn's disease and may be a useful treatment option to consider for inducing and maintaining remission.