Send to

Choose Destination
See comment in PubMed Commons below
Arch Intern Med. 2007 Feb 12;167(3):265-70.

Simple clinical predictors may obviate urgent endoscopy in selected patients with nonvariceal upper gastrointestinal tract bleeding.

Author information

Departments of Medicine and Biometry, Bioinformatics and Epidemiology, Medical University of South Carolina, 96 Jonathan Lucas Street, Charleston, SC 29425, USA.



The validated Blatchford risk score (BRS) predicts outcomes in patients with nonvariceal upper gastrointestinal tract bleeding, before endoscopy; completion of the Rockall score requires endoscopy. The aims of this study were to predict whether the modified BRS (mBRS) can predict (1) endoscopic high-risk stigmata (HRS) and (2) rebleeding and mortality.


Clinical and demographic characteristics on 1869 patients from 6 Canadian provinces were prospectively entered into the Registry for Upper GI Bleeding and Endoscopy database, recording 30-day rebleeding and mortality. The Rockall score and mBRS (hemoglobin level, hemodynamic instability, and presence of melena, liver disease, or cardiac failure; urea and syncope were not recorded) were calculated. Logistic regression was used to assess the association between an mBRS of 1 or less with HRS and with rebleeding and mortality.


The mean (SD) age of the patients was 66 (17) years, with 62% men and a mean of 2.5 comorbidities. Of the 1860 patients with 30-day rebleeding data, 334 (18.0%) rebled; 5.3% died. The mBRS was 0 in 3% and 1 or less in 9.8% of patients; HRS were seen in 31.0% of patients. An mBRS of 1 or less was associated with lower rebleeding (5% vs 19%; P<.001) and mortality (0.5% vs 5.8%; P=.003), and was significant in multivariate analysis for rebleeding (odds ratio, 0.24; 95% confidence interval, 0.12-0.48) and mortality (odds ratio, 0.12; 95% confidence interval, 0.02-0.90). The HRS were less frequent when the mBRS was 1 or less (16.9% vs 32.7%; odds ratio, 0.4; 95% confidence interval, 0.3-0.6). Patients with a low mBRS with HRS had a low rebleeding rate (3.3%) and a lower apparent benefit from endoscopic therapy.


An mBRS of 1 or less identifies approximately 10% of patients with gastrointestinal tract bleeding with a low likelihood of having HRS and a low risk of adverse outcomes. A prospective randomized study is required to examine whether this subgroup of patients presenting after hours could be discharged safely from emergency departments with arrangements for (urgent) outpatient endoscopy.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Silverchair Information Systems
    Loading ...
    Support Center