Ouanes et al developed a model for predicting death or readmission within 7 days of discharge from the ICU. This can help to identify a patient who may benefit from more aggressive management. The authors are from multiple hospitals in France and one in Tunisia.
Patient selection: adult discharged from ICU with stay > 24 hours, to a ward in the same hospital
Outcome: readmission or death within 7 days of ICU discharge, which occurred in 3%
Parameters:
(1) SAPS II score on admission to the ICU
(2) SOFA score on discharge from the ICU
(3) maximum SIRS score during last 2 days in ICU
(4) central venous catheter (CVC) during ICU stay
(5) time of day when discharged from ICU
Parameter
|
Finding
|
Points
|
SAPS II on admission
|
|
|
SOFA score on discharge
|
|
10 *
|
maximum SIRS
|
<= 2
|
0
|
|
> 2
|
40
|
CVC placed
|
no
|
0
|
|
yes
|
50
|
time of discharge
|
at night
|
75
|
|
other
|
0
|
total MIR score =
= SUM(points for all 5 parameters)
Interpretation:
• minimum score: 0
• maximum score = (165 for non-SAPS, non-SOFA) + (160 for SAPS) + (10 * (24 for SOFA)) = 565
• It is unlikely that a person would be discharged from the ICU with high SAPS or SOFA scores.
• A score > 150 had a negative predictive value for death or readmission within 7 days of 98% but positive predictive value of only 8%.
The authors also developed an equation to predict the outcome:
X =
= (0.017 * (SAPS II)) + (0.19 (SOFA)) + (0.74 if CVC placed) + (0.61 for SIRS) + (0.92 if discharged at night) – 5.59
probablility of outcome =
= 1 / (1 + EXP((-1) * X)))