OBJECTIVE To compare the relapse-preventing effect and the frequency of adverse events of olsalazine and sulphasalazine in sulphasalazine-tolerant patients with ulcerative colitis. METHODS Patients in remission, with at least two episodes of active disease during the last 5 yr, were randomized to 2 g of sulphasalazine or 1 g of olsalazine daily and were followed for 6-18 months. Relapse rates in the two groups were compared using frequency and life-table analysis. Sixty-nine patients with proctitis, 140 with left-sided colitis, and 113 with subtotal or total colitis were evaluated. RESULTS In the intention-to-treat analysis, the failure rate (relapses plus withdrawals) was 54.7% in the olsalazine and 47.2% in the sulphasalazine group. In the per-protocol analysis excluding withdrawals, 44.7% relapsed in the olsalazine and 39.3% in the sulphasalazine group. Remission curves did not differ significantly, although at all time intervals the frequency of remission was slightly higher in the sulphasalazine group (p = 0.19 in the intention-to-treat analysis and p = 0.42 in the per-protocol analysis estimated by the log-rank test). Twelve patients (of whom five had diarrhea) in the olsalazine group versus eight patients in the sulphasalazine group discontinued the study because of side effects. CONCLUSIONS The relapse-preventing effect of olsalazine and sulphasalazine in sulphasalazine-tolerant patients did not differ. Furthermore, the tolerability of olsalazine, particularly concerning diarrhea, appears to be better than previously reported.