Implementation of a standardized multi-disciplinary treatment protocol for acute GI hemorrhage

J. Finegold, R. J. Garcia-Carrasquillo, P. D. Stevens, F. Van De Mierop, P. H.R. Green, M. Battagliano, F. Meyer, R. Rosenberg, S. K. Lewis, M. Rubin, J. Chabot, B. Diamond, C. J. Ughtdale

Research output: Contribution to journalArticlepeer-review

3 Scopus citations


We compared clinical outcomes of patients with bleeding peptic ulcers who were managed by a multi-disciplinary gastrointestinal rapid response team (GIRRT) using standardized protocols with clinical outcomes of a matched cohort. Methods: Consecutive patients evaluated by the GI consult service for acute UGIB were managed by the GIRRT. The team consisted of 1 of 8 attending gastroenterologists, a GI fellow, a surgical attending, and a surgical resident. Interventional radiologists were available for consultation if needed. Pts. with acute UGIB (defined as melena and/or hematemesis and/or hematochezia +/- a bloody NG aspirate) were managed by protocol. Endoscopy (EGD) was performed within 12 hours of presentation. Ulcers were endoscopically treated if either active bleeding or a visible vessel was present and standard hemostasis protocols were followed. Endoscopic therapy was injection of 1:10,000 epinephrine followed by thermal therapy with a 10 Fr heater or bipolar probe until bleeding stopped and the vessel was flattened. After initial hemostasis, pts were stratified for rebleeding risk with the Baylor Bleeding Score(BBS). Repeat EGD was performed for clinical rebleeding (defined as a fall in hct ≥ 6% with persistent melena, recurrent hematemesis or hypotension). Pts were matched with controls obtained from our endoscopy database for age, sex, lesion site, stigmata and use of videoendoscopy. Results: From July 1 to Nov. 30, 1995, 56 patients were evaluated for acute UGIB. Thirty were found to have peptic ulcers(53%). After washing the ulcer base we found: Clean Base(26%), Flat Spot(13%), clot(20%), visible vessel(16%), and active bleeding (23%). Initial hemostasis was achieved in 100%. Historic controls and GIRRT patients(n=17) are compared below: Means PRBC(Units/pt) ICU days Hospital days GIRRT n=17 3.6 1.8 8.9 Controls n=17 3.0 4.5 11.3 p value = not assessed .09 .12 GIRRT patients at high risk for rebleeding(53%) had a 19% rebleeding rate compared to 0% in the low risk group. Conclusions: 1. GIRRT produced favorable trends in the outcomes of patients with bleeding peptic ulcers. 2. We confirm that the BBS was a good predictor of rebleeding in our population.

Original languageEnglish (US)
Pages (from-to)350
Number of pages1
JournalGastrointestinal endoscopy
Issue number4
StatePublished - 1996
Externally publishedYes

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Gastroenterology


Dive into the research topics of 'Implementation of a standardized multi-disciplinary treatment protocol for acute GI hemorrhage'. Together they form a unique fingerprint.

Cite this