Limited value of early endoscopy in the management of acute upper gastrointestinal bleeding. Prospective controlled trial.

In an 11 month period, 95 patients with acute upper gastrointestinal hemorrhage underwent early fiberoptic endoscopy. Patients were randomized into two groups before endoscopy depending on whether the results of the procedure were revealed immediately or after 4 days. No attempt was made to influence the treatment or diagnostic plan. The groups were comparable with respect to historical features surrounding the bleeding episode and the anatomic site of hemorrhage. There were no significant differences in any aspect of patient management or outcome between groups. Important changes in diagnostic or treatment plan were made after learning the results of endoscopy in only 12% of the patients, half of whom were bleeding briskly at the time of endoscopy. It is concluded that the natural history of acute upper gastrointestinal hemorrhage (for example, the spontaneous cessation of bleeding) precludes endoscopy from having a significant effect on patient management.