Endoscopy based triage significantly reduces hospitalization rates and costs of treating upper GI bleeding (UGIB): A randomized controlled trial

J. G. Lee, Samuel D Turnipseed, N. Melnikoff, R. Hsu, James D Kirk, Joseph Leung, H. Vigil, John S Rose, P. Sokolove

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


Purpose: to determine if endoscopy based triage can reduce the costs and the hospitalization rates of patients with non variceal UGIB. Methods: all stable patients with UGIB were randomized immediately after admission to be admitted as previously determined and undergo endoscopy in 24-48 hrs (control) or to undergo endoscopy in the emergency department (ED) and be discharged if it did not show active bleeding or stigmata. Outcomes were prospectively followed for 30 days. Results: 312 of 12,966 patients admitted from the ED between 9/96-9/97 had UGIB. 206 were excluded (unstable=80, varices=69, unwilling or refused=15, missed=5, other=37) and 110 were randomized. Their baseline demographic features were: Randomization Mean values Initial site of admission N Age Men Hgb ICU Intermediate Ward Control 54 59 74% 12.1 15% 35% 50% Endoscopy 56 53 70% 12.0 27% 21% 52% All p values >0.05. 64% of the emergent endoscopies were performed out of hours. The prevalence of ulcer (48 v 59%), stigmata (20 v 32%), active bleeding (13 v 14%), and hemostatic therapy (13 v 21%) were similar for the control and endoscopy groups, respectively. Preadmission endoscopy downgraded 68% and upgraded 14% of the randomized patients, allowing for immediate discharge of 26 of 56 (46%) patients. The total hospital stay was shorter in the endoscopy group, p=0.0001. No one discharged from the ED rebled or returned to the hospital. The 30 day outcomes were: Randomization Rebleed Surgery Reendoscope Readmit Death Control 6% 2% 7% 15% 4% Endoscopy 4% 2% 7% 7% 0 All p values >0.05. Although significantly more (p=0.02) control patients made unplanned visits to physicians during the follow up period, the prevalence of GI related visits was similar. The median true costs was $3,323 for the control group and $1,974 for the endoscopy group ($672 for outpatients vs $3,123 for inpatients). Conclusions: Emergent endoscopy performed in the ED can safely triage 46% of patients with UGIB for outpatient care. Although the clinical outcome of outpatient care is similar to that of inpatient care, it is significantly less expensive.

Original languageEnglish (US)
JournalGastrointestinal Endoscopy
Issue number4
StatePublished - 1998

ASJC Scopus subject areas

  • Gastroenterology


Dive into the research topics of 'Endoscopy based triage significantly reduces hospitalization rates and costs of treating upper GI bleeding (UGIB): A randomized controlled trial'. Together they form a unique fingerprint.

Cite this