A randomised controlled trial of four management strategies for dyspepsia: relationships between symptom subgroups and strategy outcome

Abstract

Background: The first step in the management of uncomplicated dyspepsia in primary care often consists of prescribing empirical therapy, but in certain cases prompt endoscopy might be preferred. Any decision is usually based on the patient's symptoms and the presumed underlying pathology that causes these symptoms.

Aim: To assess the relationship between symptom subgroups and the effect of management strategies on primary care patients with dyspepsia.

Design of study: Randomised controlled trial.

Setting: All patients presenting successively with a new episode of dyspepsia between January 1995 and November 1997.

Method: The results of four management strategies in dyspeptic primary care patients were compared and the value of subgrouping within this trial was estimated. Patients were allocated to one of either (a) empirical treatment in which therapy was based on the presented symptoms, or empirical treatment with (b) omeprazole or (c) cisapride regardless of the presented symptoms, or (d) prompt endoscopy followed by the appropriate treatment. Patients were retrospectively classified into the subgroups for each strategy using baseline data. The yield of each strategy was measured by counting the number of strategy failures in the first year.

Results: Of the 349 included patients, 326 were analysed. No statistically significant difference could be demonstrated between the strategies or between the symptom subgroups. However, patients in the reflux-like subgroup showed a trend towards a better outcome in all empirical strategies. Ulcer-like dyspepsia seemed to benefit from omeprazole. The non-specific subgroup seemed to benefit from cisapride but also had the highest proportion of strategy failure. Prompt endoscopy did not appear especially useful in any subgroup.

Conclusion: Although this study has relatively low power, we conclude that the use of symptom subgroups seems to be a sensible approach when choosing empirical therapy in dyspepsia. Patients with reflux-like symptoms seem to have the best prognosis in the first year in every strategy.