Three-hundred and twenty-five patients with endoscopically verified oesophagitis entered a double-blind, randomized multicentre study that compared 300 mg nizatidine b.d., 300 mg nocte and placebo. The 6- and 12-week treatment responses were studied. Healing was defined as complete epithelialization of all oesophageal lesions. The healing rates were 40% in the 300 mg nizatidine b.d. group, 30% in the 300 mg nocte group and 26% in the placebo group at 6 weeks. The corresponding figures after 12 weeks of treatment were 50%, 44% and 34%, respectively. The healing rates were significantly different (P less than 0.05) between the high-dose nizatidine group and placebo only, both at 6 and 12 weeks. Despite a trend at both 6 and 12 weeks in favour of 300 mg nizatidine nocte compared to placebo, this was not significantly different. The most important factor for the outcome, apart from the treatment group, was the pre-entry severity of oesophagitis. The differences observed between treatment groups in healing rates, symptomatic relief, and antacid consumption appear to result mainly from the patients with moderate and severe oesophagitis upon entry. Nizatidine (300 mg) b.d. appeared to be safe and effective in the treatment of reflux oesophagitis.