[Critical study of gastroduodenal complications of corticotherapy].

Contrary to a still widely held belief, there is no explicative or clinical evidence that corticosteroid therapy has undesirable effects on the digestive tract causing, in particular, gastric or duodenal ulcers. From a meta-analysis of all published randomized, double-blind trials comparing corticosteroids with placebos, it appeared that the prevalence of peptic ulcers was 0.4% in the steroid group (n = 3335) and 0.3% in the placebo group (n = 3267), the difference being non-significant. Although a weak cause-effect relationship cannot be definitely excluded, these data demonstrate that it is unnecessary in practice to prescribe preventive anti-ulcer treatments to all patients receiving corticosteroids.