IT is known that adrenocorticosteroid therapy is frequently complicated by the appearance or reactivation of peptic ulcers,1 and that these ulcers are often associated with bleeding and perforation.2 3 4 The basis for this knowledge is difficult to define. Although there are many anecdotal reports of the development of peptic ulcer in patients receiving adrenocorticosteroid (steroid) treatment that go back to the early days of cortisone therapy5 6 7 and several series of patients with a high incidence of ulcers,2 3 4 , 8 , 9 no definitive description of this relation can be found, and no objective confirmation of this association exists. A few reports that failed to find . . .