The evolution of stigmata of hemorrhage in bleeding peptic ulcers: a sequential endoscopic study.

BACKGROUND AND STUDY AIMS Stigmata of hemorrhage in bleeding peptic ulcers have prognostic characteristics. In the present study, the evolution of these stigmata was studied prospectively using daily endoscopic examinations. PATIENTS AND METHODS From January 1989 to October 1989, 778 consecutive patients with bleeding peptic ulcers underwent endoscopy within 24 hours of admission. The bleeding peptic ulcers were assigned by three endoscopists to five categories, those with: a) active bleeding, b) a nonbleeding visible vessel, c) adherent clot, d) dot, or e) a clean base. Actively bleeding ulcers were treated by epinephrine injection. Ulcers with nonbleeding visible vessels, adherent clots, or dots were left untreated. Daily endoscopic examinations were carried out for three subsequent days, or until the ulcer base became clean. RESULTS On day 0, there were 56 actively bleeding ulcers (7%), 62 ulcers with visible vessels (8%), 104 with adherent clots (13%), 182 with flat dots (23%), and 374 with a white base (48%). On the subsequent three days, 24 of 62 ulcers with visible vessels (39%), 30 of 104 with adherent clots (29%), 24 of 182 with dots (13%), and 19 of 374 with a clean base (5%) on day 0 re-bled endoscopically or clinically, or both. The overall rebleeding risk was 9.9%, 4.9%, and 2.7% on days 1, 2, and 3, respectively. CONCLUSIONS Stigmata of hemorrhage in bleeding peptic ulcers are predictive of rebleeding. They represent intermediate phases in the evolution of bleeding vessels into clean-based ulcers. The associated rebleeding risk diminishes as the vessel disappears from the ulcer base.