[Skip to Content]
[Skip to Content Landing]
Article
July 25, 1990

Endoscopic HemostasisAn Effective Therapy for Bleeding Peptic Ulcers

Author Affiliations

From the Clinical Trials Unit and the Departments of Medicine and Biomathematical Sciences, Mount Sinai School of Medicine, New York, NY (Drs Sacks and Chalmers, Ms Berrier, and Mr Pagano); the Technology Assessment Group and the Department of Health Policy and Management, Harvard School of Public Health, Boston, Mass (Dr Chalmers); the Boston Veterans Administration Hospital (Dr Chalmers); and the Division of Gastroenterology, University Hospital, Lausanne, Switzerland (Dr Blum).

From the Clinical Trials Unit and the Departments of Medicine and Biomathematical Sciences, Mount Sinai School of Medicine, New York, NY (Drs Sacks and Chalmers, Ms Berrier, and Mr Pagano); the Technology Assessment Group and the Department of Health Policy and Management, Harvard School of Public Health, Boston, Mass (Dr Chalmers); the Boston Veterans Administration Hospital (Dr Chalmers); and the Division of Gastroenterology, University Hospital, Lausanne, Switzerland (Dr Blum).

JAMA. 1990;264(4):494-499. doi:10.1001/jama.1990.03450040090037
Abstract

We performed a meta-analysis of 25 randomized control trials that compared endoscopic hemostasis with standard therapy for bleeding peptic ulcer. For recurrent or continued bleeding, the mean rate in control patients was 0.39, and the pooled rate difference, or reduction due to therapy, was 0.27 ±0.15 (95% confidence interval) (69% relative reduction). For emergency surgery, the mean rate in control patients was 0.26, and the pooled rate difference was 0.16 ± 0.05 (62% relative reduction). Most important, for overall mortality, the mean rate in control patients was 0.10, and the pooled rate difference was 0.03 ±0.02 (30% relative reduction). The effects were greatest in patients with spurting or visible blood vessels and equivocal when the ulcer showed only signs of recent bleeding. We conclude that endoscopic hemostasis is clearly effective but that data were insufficient for direct comparisons between modalities. Randomized control trials to compare the different modes of endoscopic therapy should continue.

(JAMA. 1990;264:494-499)

×