Description

Kim et al developed a model for predicting perioperative blood transfusion for a patient undergoing colorectal, hepatic or pancreaticobiliary surgery. This can help to identify a patient who may benefit from preoperative interventions and blood conservation during the procedure. The authors are from the Johns Hopkins University.


 

Patient selection: undergoing colorectal, hepatic or pancreaticobiliary surgery

 

Outcome: perioperative blood transfusion

 

Parameters:

(1) age

(2) gender

(3) race

(4) preoperative hemoglobin

(5) preoperative INR

(6) type of surgical procedure

(7) Charlson comorbidity index (CCI)

 

Parameter

Finding

Points

age

<= 65 years of age

0

 

> 65 years of age

13

gender

female

0

 

male

4

race

White

0

 

Black

14

 

Asian

19

 

other

5

preoperative hemoglobin

> 12.0 g/dL

0

 

11.1 to 12.0 g/dL

24

 

10.1 to 11.0 g/dL

42

 

9.1 to 10.0 g/dL

58

 

8.1 to 9.0 g/dL

74

 

<= 8.0 g/dL

100

preoperative INR

<= 1.2

0

 

> 1.2

27

type of procedure

colon surgery

0

 

minor hepatectomy

4

 

rectal surgery

10

 

major hepatectomyh (hemi-hepatectomy or more)

16

 

other pancreatectomy

23

 

duodenopancreatectomy

30

Charlson comorbidity score

<= 3

0

 

> 3

19

 

total score =

= SUM(points for all 7 parameters)

 

Interpretation:

• minimum score: 0

• maximum score: 212

• The higher the score the greater the chances of a perioperative blood transfusion.

 

Total Score

Probability of Perioperative Blood Transfusion

< 7.7

< 10%

7.7 to 49

(0.00489 * ((points)^2)) + (0.2089 * (points)) + 8.102

49 to 117

(0.7526 * (points)) - 6.323

117 to 164

(-0.00437 * ((points)^2)) + (1.543 * (points)) - 40.51

> 164

> 95%

 

Performance:

• The area under the ROC curve was 0.76.

 


To read more or access our algorithms and calculators, please log in or register.