Models of Guo et al for Predicting Massive Transfusion in a Trauma Patient: LASSO Model

Description

Guo et al reported 2 models for predicting massive transfusion in a trauma patient. One model is based on LASSO and the second was based on vital signs alone. The authors are from Chinese PLA General Hospital Beijing and Tsinghua University in China.

Patient selection: trauma patient

Parameters:

(1) heart rate in beats per minute

(2) pulse pressure in mm Hg (systolic pressure minus diastolic pressure)

(3) serum base excess in mmol/L

(4) hemoglobin in g/L

(5) displaced pelvic fracture

(6) CT or FAST positive

Parameter	Finding	Beta Coefficient
heart rate		0.013 * (heart rate)
pulse pressure		-0.014 * (pressure)
serum base excess		-0.104 * (base excess)
hemoglobin		-0.048 * (hemoglobin)
displaced pelvic fracture	no	0
	yes	0.757
examinations	CT positive	0.484
	FAST positive	0.484
	both positive	0.484
	both negative	0

value of X =

= SUM(points for all of the beta-coefficients) + 2.419

probability of massive transfusion =

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

Performance:

• The area under the ROC curve is 0.89.

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

Models of Guo et al for Predicting Massive Transfusion in a Trauma Patient: LASSO Model

Description

Specialty