Peptic Ulcer Associated with Corticosteroid Therapy: Serial Roentgenographic Studies

Abstract

During a period of two and one-half years, 77 patients who were candidates for or were receiving steroid therapy, were observed for peptic ulcer. Most of them were examined roentgenologically at three-month intervals. This made it possible to correlate ulcer incidence with dose, duration of therapy, and effect of individual drugs, as well as to add to the very sparse information upon the behavior of ulcers during continuous steroid administration. Almost all of the patients had systemic lupus erythematosus and required large amounts of steroids. The incidence of peptic ulcer in this treated group was compared with the incidence in a similar group of patients not receiving these hormones. Since there are relatively few reports of roentgenologic surveys of patients treated with steroids, and none with such extensive serial studies while on high dosage, it was thought that presentation of our findings and comparison with previous data would be of interest. Up to 1954, Sandweiss (1) and Wollaeger (2) were able to collect from the literature only 50 to 55 instances of peptic ulcer occurring during steroid therapy. Since then the incidence has been variously reported as 0 to 31 per cent. When cortisone alone was available, several reports indicated an incidence of less than 7.8 per cent (3–7). With the use of prednisone, although a few investigators (8, 9) found complicating peptic ulcer in less than 5 per cent, the figures of others ranged from 12.5 to 31.0 per cent (10–18). Among the latter, Kern, Clark, and Lukens (12) and Kammerer et al. (17, 18) showed a rather clear-cut increase in ulcer incidence with increasing steroid dosage. The data of Kammerer and his associates are especially striking, both as to total ulcer incidence and dosage correlation. They report an overall incidence of 31 per cent. When 10 mg. of prednisone or its equivalent was given, there was an almost tenfold increase in ulcer incidence over that following lower doses. All of their patients had received steroids for six months or more, and a single roentgenologic study was made of each patient. Prolongation of therapy beyond six months did not affect ulcer incidence. Gedda and Moritz (16) found an increased risk of peptic ulcer with increasing length of treatment. Others failed to make such a correlation. Since we employed roentgen methods of diagnosis, we were particularly interested in studies using similar diagnostic criteria. Bollet, Black, and Bunim (10), Evans (15), Gedda and Moritz (16), and Kammerer et al. (17, 18) reported the occurrence of ulcer in 16 to 31 per cent of their series, and these authors in each instance remarked upon the frequency with which the patients were asymptomatic or had symptoms atypical for ulcer in the face of an objectively demonstrated lesion. The only radiologic survey showing an extremely low incidence of ulcer is that of Meltzer et al. (9).