Summary
The widely differing mortality rates of severe gastroduodenal hemorrhage reported
in the literature (10-30 %), are due to very inhomogeneous patient groups. The purpose
of this study was therefore to rank various clinical and endoscopical factors by giving
them points from 0 to 6, both to establish comparable groups and to use them as prognostic
parameters for a potential fatality rate.
One hundred and ninety-three patients with severe gastroduodenal bleeding, verified
at emergency gastroscopy immediately after admission, were admitted to this prospective
study.
The statistical assessment of the prognosis for these 8 factors revealed a linear
correlation between increasing score and mortality rate for the following risk factors:
patient age, activity and intensity of hemorrhage, type and number of associated illnesses,
various therapeutic procedures. Only the type of the source and the site of the hemorrhage
did not correlate well.
By adding up the points of all risk factors for every patient we calculated the overall
score and established a correlation to the mortality rate: A definite statistical
correlation was demonstrated between increasing score and fatal outcome. If a patient
with a score of less than 20 points did survive in 100 %, the mortality rate increased
linearly to 83.3 % in patients with a score of 40 points.
Using this scoring system it is possible to establish comparable groups of patients
- which seems indispensable for a critical examination of various therapeutic procedures.
Furthermore, this score can serve as a predictor of the probability of a fatal outcome
shortly after patient admission.
Key words:
Gastroduodenal hemorrhage - Clinical and endoscopic factors - Mortality rate
