SUMMARY Sixty patients with gastric ulcers were treated for four weeks with either 1 g cimetidine per day or with identical tablets containing lactose. The healing rate, assessed by endoscopy, was 23 out of 35 (66%) in the patients given cimetidine and 13 out of 25 (52%) in those given placebo. The difference between the groups is not significant. During each of the four weeks of the study the cimetidine group experienced significantly fewer attacks of pain and consumed less antacids than the placebo treated patients.The effectiveness of the histamine H2 antagonist cimetidine in duodenal ulcer is now well known. Its value in gastric ulcer has yet to be established. We report the results of a double blind trial of four weeks' treatment with cimetidine 1 g daily in the management of gastric ulcer.
MethodPatients admitted to the trial were ambulant and had been shown to have a gastric ulcer by fibreoptic gastroscopy within the previous four days. None had previously received cimetidine. At the beginning only patients considered unsuitable for treatment with carbenoxolone were included. These were either aged over 60 years, had a history of cardiovascular disease, or had already been unsuccessfully treated with carbenoxolone. After encouraging preliminary results, however, with 80 % healing of the first ulcers studied (Ciclitira et al., 1976), it was decided to include patients who previously would have been treated with carbenoxolone. Any ulcer proximal to the pylorus was included in the study and all were brushed and biopsied to exclude malignancy.Patients were allocated at random to receive either cimetidine 200 mg tds and 400 mg at night for four weeks, or identical tablets containing lactose. Both groups were supplied with unlabelled antacid tablets whose buffering capacity is 15-1 mmol per tablet (Rennies, Nicholas Laboratories), to be taken whenever necessary for relief of ulcer pain. All other 'Drummond Research Fellow 1977-79. "Present address: Wellcome Research Fellow, Department of Gastroenterology, St Bartholomew's Hospital, London EC1. Received for publication 12 February 1979 therapy was forbidden. Symptoms in the week before entering the trial were assessed by asking the patient to recall how many attacks of pain he had suffered. Subsequently all patients recorded on diary cards the number of attacks of pain suffered and the number of antacid tablets consumed. A return tablet count was made at each clinical visit to check that treatment had been correctly taken.The patients were seen in outpatients after two weeks and at the end of the four week trial period. At these visits haemoglobin, white cell count, liver function tests, urea and electrolytes estimations were performed. At the end of the trial the endoscopy was repeated.
ResultsSeventy patients entered the trial. Ten had to be withdrawn for various reasons including the subsequent discovery of malignancy in the biopsy (six) or default (two). In two cases (both on placebo) haemorrhage occurred within 48 hours of entry to the trial. These case...