THE CHOICE OF ULCER HEALING AGENT INFLUENCES DUODENAL ULCER RELAPSE RATE AND LONG‐TERM CLINICAL OUTCOME

Publisher Website
Google Scholar
Abstract
The mode of pharmacological healing influences subsequent ulcer relapse. Duodenal ulcer relapse following healing and withdrawal of active therapy has been studied in 18 prospectively randomised comparative studies involving cimetidine and alternative healing agents. Most studies (eleven of the 18) showed comparative agents with a lower (greater than 10%) incidence of relapse than cimetidine while a minority (seven of the 18) showed equal (+/- 10%) relapse. The probability of this distribution occurring by chance is less than 1:1000 (p = 0.0005, sign test). In three of the 11 published comparative studies the relapse rate was significantly higher in those who had previously received cimetidine (p less than 0.05, two-sided significance test). This was despite the fact that most of the individual studies were small and required a substantial difference in results for the 5% significance level to be reached. Paired comparative incidences of relapse at six months were converted to relapse rates (percentage of healed population relapsing monthly), average rates for comparative agents were 8.05 +/- 2.7% while those for cimetidine were 16.9 +/- 4.4% (means +/- SEM, n = 5); cimetidine relapse rates derived from non-comparative trials were 17.0 +/- 1.9% (means +/- SEM, n = 7). Higher relapse rates translate into larger numbers of patients at risk and in need of active therapy. These results are of biological and clinical significance. Differences in relapse must reflect beneficial effects of one group of agents on the ulcer diathesis or adverse effects of the comparator, cimetidine. Each alternative reflects negatively on cimetidine as an antiulcer medication; however, confirmation of adverse effects would force radical review of most current therapy of ulcer disease. Research into ulcer therapeutics must emphasise both the relapse and the healing potential of pharmacologic agents.