Meta-analysis of the effect of placebo on the outcome of medically treated reflux esophagitis.

Abstract

To ascertain the placebo-induced effect in the treatment of reflux esophagitis, we reviewed all the English-language literature concerning the results of placebo-controlled trials of erosive/ulcerative esophagitis from 1976 to 1990. Twenty-two studies fulfilled our meta-analytic criteria. After 4 to 8 weeks of treatment, active drugs (cimetidine, ranitidine, nizatidine, omeprazole, metoclopramide, sucralfate) were significantly more effective than placebo in the healing of esophagitis, with a pooled rate difference (PRD) of 0.22 in favor of the active drug, an odds ratio (OR) of 2.57 (confidence interval (CI) = 2.0-3.3). Pooled mean healing rate (+/- SD) with the active drug was 47.3 +/- 24.0%, as compared with 26.8 +/- 18.0% obtained with placebo after 4 to 8 weeks of treatment. With regard to symptomatic response, complete disappearance of symptoms was observed in an average of 31.6% active-treated patients and in 11.8% of placebo-treated patients, respectively. The PRD was 0.20, and the OR 2.25 (CI = 1.65-3.06). The incidence of side effects was not statistically different for the two treatment groups. Placebo is a relatively inactive drug in the short-term treatment of erosive ulcerative reflux and does not appear to change the natural history of the disease.