Elsevier

The Lancet

Volume 373, Issue 9657, 3–9 January 2009, Pages 42-47
The Lancet

Articles
Outpatient management of patients with low-risk upper-gastrointestinal haemorrhage: multicentre validation and prospective evaluation

https://doi.org/10.1016/S0140-6736(08)61769-9Get rights and content

Summary

Background

Upper-gastrointestinal haemorrhage is a frequent reason for hospital admission. Although most risk scoring systems for this disorder incorporate endoscopic findings, the Glasgow-Blatchford bleeding score (GBS) is based on simple clinical and laboratory variables; a score of 0 identifies low-risk patients who might be suitable for outpatient management. We aimed to evaluate the GBS then assess the effect of a protocol based on this score for non-admission of low-risk individuals.

Methods

Our study was undertaken at four hospitals in the UK. We calculated GBS and admission (pre-endoscopy) and full (post-endoscopy) Rockall scores for consecutive patients presenting with upper-gastrointestinal haemorrhage. With receiver-operating characteristic (ROC) curves, we compared the ability of these scores to predict either need for clinical intervention or death. We then prospectively assessed at two hospitals the introduction of GBS scoring to avoid admission of low-risk patients.

Findings

Of 676 people presenting with upper-gastrointestinal haemorrhage, we identified 105 (16%) who scored 0 on the GBS. For prediction of need for intervention or death, GBS (area under ROC curve 0·90 [95% CI 0·88–0·93]) was superior to full Rockall score (0·81 [0·77–0·84]), which in turn was better than the admission Rockall score (0·70 [0·65–0·75]). When introduced into clinical practice, 123 patients (22%) with upper-gastrointestinal haemorrhage were classified as low risk, of whom 84 (68%) were managed as outpatients without adverse events. The proportion of individuals with this condition admitted to hospital also fell (96% to 71%, p<0·00001).

Interpretation

The GBS identifies many patients presenting to general hospitals with upper-gastrointestinal haemorrhage who can be managed safely as outpatients. This score reduces admissions for this condition, allowing more appropriate use of in-patient resources.

Funding

None.

Introduction

Upper-gastrointestinal haemorrhage is a frequent cause of acute admission to hospital, with an incidence in the UK of 103–172 per 100 000 adults per year.1, 2 The severity of the disorder varies from mild coffee-ground vomiting to exsanguination. However, most patients do not need endoscopic treatment, surgery, or blood transfusion and do not rebleed or die.1, 3 Individuals presenting with upper-gastrointestinal haemorrhage have traditionally been admitted for a period of observation, with or without endoscopy.

Admission and endoscopy on the next available list is recommended in the 2002 British Society of Gastroenterology guideline for people with mild-to-moderate upper-gastrointestinal haemorrhage,4 although very low-risk young people with a minor bleed and without haemodynamic compromise can be discharged without endoscopy. We know from our experience and in other hospitals that some clinicians use their judgment informally to avoid admittance of individuals they view as being at low risk. However, objective identification of such patients with clinical confidence is sometimes difficult.

Several risk assessment and scoring systems for upper-gastrointestinal haemorrhage have been developed in an attempt to stratify risk for poor outcome.2, 5, 6, 7, 8, 9, 10, 11, 12 However, most, including the widely used Rockall score,3 include endoscopic findings; therefore, many patients are kept in hospital until this procedure is undertaken. Although many hospitals in the UK have an emergency endoscopy rota, this facility is usually for individuals with major haemorrhage only, with others waiting until the next day or longer for a semi-elective procedure. Furthermore, non-emergency endoscopy is unavailable at weekends in many hospitals. An abbreviated pre-endoscopy admission Rockall score, which excludes endoscopic findings, is sometimes used, but this measure has not been fully validated.3

In a previous report from Glasgow, UK, logistic regression was used to derive the Glasgow-Blatchford bleeding score (GBS; table 1), which is used to predict either a patient's need for hospital-based intervention (blood transfusion, endoscopic treatment, or surgery) or death.5 The score was derived from data of 1748 people presenting with upper-gastrointestinal haemorrhage but was only validated locally in a few affected individuals presenting to three Glasgow hospitals, not including the Glasgow Royal Infirmary. It is based on simple variables from a patient's history, examination, and laboratory results. A GBS score of 0 fulfils low-risk criteria (panel), which seems to identify people at very low (0·5%) risk of needing intervention, as described above.5

The aim of our study was to assess and externally validate the GBS in four large general hospitals in Scotland and England. We also prospectively looked at the effect of the introduction of GBS low-risk criteria on accident and emergency (A&E) departments, with the intention to avoid admission for patients assessed as low risk.

Section snippets

Data collection

We divided our study into two phases. In phase one, we obtained data prospectively from consecutive patients presenting with upper-gastrointestinal haemorrhage over a 12-month period at Royal Cornwall Hospital, Truro, for 6 months at Glasgow Royal Infirmary, Glasgow, and over 3 months at Ninewells Hospital, Dundee, and retrospectively for 3 months at University Hospital of North-Tees, Stockton. We defined upper-gastrointestinal haemorrhage as haematemesis, coffee-ground vomit, or melaena. We

Results

From the four study centres, a total of 676 patients were included in phase one. Table 2 outlines demographic characteristics and outcomes for these people.

19 individuals had data missing for measurement of admission Rockall score and 27 had omissions for GBS. Of those with complete data, GBS was 0 (low-risk criteria met) in 105 (16%) and admission Rockall score was 0 in 184 (28%). The GBS low-risk group consisted of 27 people (12%) from Truro, 17 (17%) from Stockton, 36 (17%) from Glasgow, and

Discussion

Our findings show that simple GBS low-risk criteria can identify a significant proportion of individuals presenting with upper-gastrointestinal haemorrhage who are suitable for outpatient management. Furthermore, use of these criteria in A&E departments leads to a reduction in admissions for this disorder, with no apparent deleterious effects on patients' care.

Although most scoring systems for upper-gastrointestinal haemorrhage incorporate endoscopic findings, outcomes of an audit by the

References (24)

  • O Blatchford et al.

    Acute upper gastrointestinal haemorrhage in west of Scotland: case ascertainment study

    BMJ

    (1997)
  • TA Rockall et al.

    Risk assessment after acute upper gastrointestinal haemorrhage

    Gut

    (1996)
  • Cited by (279)

    • Scoring systems for risk stratification in upper and lower gastrointestinal bleeding

      2023, Best Practice and Research: Clinical Gastroenterology
    View all citing articles on Scopus
    View full text