Model of Modanlou et al for Identifying a Patient Who May Use Blood Products During Liver Transplant Surgery

Description

Modanlou et al developed model which can help to identify a patient who may be transfused with more blood during liver transplant surgery. More blood can be made available if greater blood loss is expected. The authors are from Saint Louis University in Saint Louis, Missouri.

Blood usage during liver transplant surgery: Blood was delivered to the operating room in boxes containing 10 units of packed RBCs and 10 units of plasma.

Outcome: using more than one 10 unit "box" of packed RBCs

Models:

(1) simple point score

(2) logistic regression model

Parameters:

(1) age of the liver donor in years

(2) serum creatinine of the recipient in mg/dL

Parameter	Finding	Points	Beta
age of liver donor	<= 50 years	0	0
	> 50 years	1	1.02
recipient's serum creatinine	<= 1.3 mg/dL	0	0
	> 1.3 mg/dL	1	1.34

simple score =

= SUM(points for the 2 parameters)

Simple Score	Percent Using More Blood
0 or 1	44%
2	80%

X for logistic regression model =

= SUM(beta coefficients) - 0.76

probability of using more blood =

= 1 / (1 + EXP((-1) * X))

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

Description

Specialty

ICD-10