Prokinetic agents are being used increasingly in medical therapy for gastro-oesophageal reflux disease (GERD). This study examined the effect of 10 mg q.d.s., oral cisapride, or placebo, taken for 12 weeks, on 48 patients with symptoms and endoscopic evidence of GERD. Objective evaluation of benefit was obtained by endoscopy and biopsy, oesophageal manometry, acid reflux provocation test and 24-h oesophageal pH monitoring. Cisapride significantly increased lower oesophageal sphincter pressure (P = 0.003) against baseline and also against placebo, in patients (n = 9) with an hypotensive lower oesophageal sphincter pressure (P < 0.01). The frequency of dyspeptic symptoms was significantly improved in the cisapride group (P = 0.03). Antacid intake, global evaluation of symptoms and a VAS score for symptoms were all better than placebo but failed to reach significance (global evaluation by patients, P = 0.07). Overall, there was no significant improvement in oesophagitis at either 6 weeks (P < 0.05 > 0.3) or 12 weeks (P = 0.07). However, if patients with grades I and II oesophagitis at entry were excluded, cisapride had a significantly greater effect than placebo, 6 weeks (P = 0.05), 12 weeks (P = 0.04). In those with oesophageal ulceration, cisapride was significantly more effective than placebo in inducing healing. Gastro-oesophageal reflux was very variable on both 24-h pH monitoring and acid reflux provocation test. In spite of a 50% decrease in acid exposure on 24-h pH monitoring (cisapride group, mean % pH < 4 day: entry 18.9%, 12 weeks 9.6%), there were no significant intra- or intergroup differences for percentage of time < pH 4, or frequency and duration of episodes, neither pre- or post-prandially, day or night, except for the number of post-prandial episodes during acid reflux provocation tests, which decreased significantly more with cisapride than with placebo (P < 0.05). Thus, oral cisapride when taken for 12 weeks promoted healing of oesophagitis and improved symptoms in patients with GERD; although an increase in lower oesophageal sphincter pressure was observed and a reduction in acid reflux was measured, no significant decrease of acid exposure was seen.