Description

Wade et al developed a model for predicting massive transfusion in a trauma patient. A patient requiring massive transfusion can be identified soon after coming to the hospital. The authors are from the U.S. Army Institute for Surgical Research.


 

Parameters seen on admission:

(1) systolic blood pressure in mm Hg

(2) heart rate in beats per minute

(3) arterial pH

(4) hematocrit in percent

 

X =

=(0.0158 * (heart rate)) - (0.0113 * (systolic blood pressure)) – (0.0932 * (pH)) – (0.0395 * (hematocrit)) + 32.8738

 

where:

• In the published abstract the factor for pH is –0.0932. In Maegle et al the factor is +0.0932. The risk associated with pH should be with acidosis.

• When the equation is modeled as written the results do not make sense. A modified equation with (a) the heart rate as a negative weighitng and (b) each weighting multiplied by 10 give better results.

 

probablility of massive transfusion =

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

 

Interpretation:

• The cut-off was a probability >= 50%.

• The patients who required transfusion tended to have a higher Injury Severity Score (ISS).

 

Performance:

• The sensitivity was 87% and specificity 53%.

• The positive predictive value was 75% and negative predictive value 72%.

 


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