Parameters:
(1) sepsis
(2) resolution of renal failure
(3) respiratory failure
(4) oliguria
(5) hospital service (medical vs surgical)
|
Parameters
|
Finding
|
Points
|
|
sepsis
|
present
|
1
|
|
|
absent
|
0
|
|
resolution of renal failure
|
renal failure resolved
|
1
|
|
|
renal failure not resolved
|
0
|
|
respiratory failure
|
present
|
1
|
|
|
absent
|
0
|
|
type of renal failure
|
oliguric
|
1
|
|
|
nonoliguric
|
0
|
|
hospital service
|
medical
|
1
|
|
|
surgical
|
0
|
where:
• Coding of patients by hospital services other than medicine or surgery is not stated. A person on the surgical service is more likely to have a reversible cause occurring post-operatively.
• I assume that anuric renal failure would be coded as oliguric.
X =
= (1.804 * (points for resolution)) – (1.106 * (points for sepsis)) – 0.227
Y =
= 0.123 – (0.841 * (points for respiratory failure)) – (1.358 * (points for sepsis)) – (0.565 * (points for oliguria)) – (0.671 * (points for service))
probability of survival =
= 1 / (1 + EXP((-1) * X))
probability of recovery of renal function =
= 1 / (1 + EXP((-1) * Y))
Performance:
• The prediction of outcome had a sensitivity of 75% and specificity of 80%.
Limitations:
• The study was performed more than 15 years ago, and survival is likely to have improved in the interval. However, the scores reflect multiple organ failures, which are still risk factors for poor outcome.
