BACKGROUNDS/AIMS: The aim of this study was to prospectively compare the efficacy of an open treatment with rabeprazole (R) and cimetidine (C) for the prevention of symptomatic relapse of GERD. METHODS: Ninety one patients with GERD who had become symptom-free after their initial treatment were asked to choose either R (10 mg/day) or C (1200 mg/day) for their a maintenance therapy. They visited the outpatient clinic every 4 weeks to evaluate their symptoms. RESULTS: The median and maximum follow-up periods were 8 and 76 weeks, respectively, in the R group and 8 and 68 weeks, respectively, in the C group. Treatment failure rates were 6.5% in the R group and 48.9% in the C group. Beside the failure of treatment, the reasons for the patients' request for changing their medications were mainly the expensive cost in the R group and the unsatisfactory effect in the C group. The mean symptom-free period in the R and C groups were 69.8 weeks and 31.3 weeks, respectively (p<0.001). The relative hazard ratio of symptom relapse was 7.55 (p=0.002) in group C as compared with group R. The proportion of patients without symptom at 24 weeks was 90.6% in the R group and 57.4% in the C group. The proportion of patients without symptoms at 52 weeks was 90.6% in the R group and 0% in the C group. CONCLUSIONS: As a maintenance therapy, rabeprazole is far more effective than cimetidine for preventing the relapse of GERD symptoms.