Abstract

A refractory duodenal ulcer was arbitrarily defined as one that had failed to heal completely after treatment with cimetidine 1 g daily for three months. Of 66 patients with refractory duodenal ulcer, healing eventually occurred in 37 patients, after treatment for an average of 7.4 months. But 28 patients did not heal despite treatment for an average of 9.4 months; and one patient defaulted. In 41 patients the daily dose of cimetidine was increased to 2 g: the ulcers in 31 patients healed. In eight patients the daily dose was increased to 3 g and healing occurred in four patients. Eighteen patients required admission on 22 occasions because of severe symptoms despite treatment. Nine patients underwent surgery but in five the results were poor. Differences in clinical and endoscopic features between refractory and non-refractory ulcer patients were small. Acid and pepsin secretion were similar and gastrin concentrations normal. Blood levels of the drug and suppression of acid secretion were both satisfactory. Identification of refractory ulcer patients at the start of treatment was therefore not possible. Refractoriness could occur at any time during the course of the disease, previous treatment with cimetidine often having resulted in rapid healing, but subsequent relapses were also usually refractory. The cause of refractoriness remains unknown and the rather poor results of surgery in this series suggests that optimal management of these patients remains to be determined. Refractoriness probably indicates a changed natural history of the disease and in some patients a more poor prognosis.