Upper GI bleed: Patients with combination of haematemesis and melaena had 50 % chance of suffering from peptic ulcer.
|
|
|
Clinical bottom line (level 4)
-
Half of patients admitted with haematemesis or melaena had a peptic ulcer, 10% have varices and 3% had a Mallory-Weiss tear.
-
Patients with haematemesis and no melaena were slightly less likely to have a peptic ulcer. The combination made a peptic ulcer slightly more likely.
-
Black haematemesis made varices slightly less likely.
-
Black haematemesis made a Mallory-Weiss tear unlikely.
|
|
Wara and Stodkilde:
Scandinavian Journal of Gastroenterology
1985;
20:
72-78
|
Expires
November 2002
|
The study
Setting: university hospital, Denmark
539 patients
(aged
1 day to 96 years; median 64,
64%
male)
admitted with haematemesis or melaena
Independent ?blinded
reference standard, applied in
all
patients from a
consecutive appropriate
spectrum.
Reference standard:
Diagnostic test:
bleeding pattern before admission
- black haematemesis: vomiting of altered blood appearing as coffee grounds
- red haematemesis: vomiting of red blood or clots
- melaena: passage of black shiny stool or noted on rectal palpation
The evidence
pre-test probability of peptic ulcer:
51.21%,
(95% CI:
47% to
55.43%)
pre-test probability of varices:
10.95%,
(95% CI:
8.3% to
13.58%)
pre-test probability of Mallory-Weiss tear:
3.3%,
(95% CI:
1.82% to
4.86%)
| diagnostic test |
peptic ulcer |
no peptic ulcer |
LR (95% CI) |
post-test probability |
| black haematemesis |
34 |
75 |
0.43
(0.30 to
0.62)
|
31% |
| black haematemesis with melaena |
50 |
20 |
2.4
(1.5 to
3.9)
|
71% |
| melaena only |
56 |
42 |
1.3
(0.88 to
1.8)
|
57% |
| red haematemesis |
49 |
71 |
0.66
(0.48 to
0.91)
|
41% |
| red haematemesis with melaena |
87 |
55 |
1.5
(1.1 to
2.0)
|
61% |
| total |
276 |
263 |
| diagnostic test |
varices |
no varices |
LR (95% CI) |
post-test probability |
| red haematemesis with or without melaena |
45 |
217 |
1.7
(1.4 to
2.0)
|
17% |
| black haematemesis or melaena |
14 |
263 |
0.43
(0.27 to
0.69)
|
5% |
| total |
59 |
480 |
| diagnostic test |
Mallory-Weiss tear |
no Mallory-Weiss tear |
LR (95% CI) |
post-test probability |
| red haematemesis with or without melaena |
18 |
244 |
2.1
(2.0 to
2.3)
|
7% |
| black haematemesis or melaena |
0 |
277 |
0.0
(0.0 to
0.29)
|
0% |
| total |
18 |
521 |
Comments
- Endoscopy failed to establish a cause in 13% of cases (95% CI: 10% to 16%).
Citation
-
Wara
P,
and
Stodkilde
H:
bleeding pattern before admission as guideline for emergency endoscopy.
Scandinavian Journal of Gastroenterology
1985;
20:
72-78
Search Terms:
?
Contributor: Chris Ball and Musab Hayatli,
November 1999
Reviewer:
Clinical Question.
| Patient |
bloody vomit, black stool |
| Intervention or Exposure |
report of symptoms, colour |
| Outcome |
source of bleeding |
|
|