Predicting massive blood transfusion using clinical scores post-trauma.

BACKGROUND AND OBJECTIVES: Early prediction of massive transfusion (MT) post-trauma may reduce mortality by earlier delivery of blood products. A clinical prediction tool (PWH score) for this purpose was developed at the Prince of Wales Hospital, Hong Kong. The aims of this study were to apply this tool to major trauma patients in Victoria, Australia and compare the score to the Assessment of Blood Consumption (ABC) score and the Trauma-Associated Severe Haemorrhage (TASH) score.
METHODS: A retrospective review of patients entered into the The Alfred Trauma Registry between January 2006 and December 2009 was conducted. The performance of the PWH score to predict MT defined by 5 units of packed red blood cells in 4 h was compared with the ABC and TASH scores. Included patients presented to the Emergency & Trauma Centre from the scene and had had complete datasets with respect to the components of the three scores.
RESULTS: There were 1234 patients included in the study with 195 (15·8%) receiving a MT and an overall mortality of 14·0%. The PWH score had an area under the receiver operating characteristics (ROC) curve of 0·842 (95% CI: 0·820-0·862). The area under the ROC curve of the PWH score was significantly less than that of the TASH score (χ(2)=19·8, P<0·001) and significantly greater than that of the ABC score (χ(2)=9·3, P=0·002).
CONCLUSIONS: The PWH score performs with similar accuracy when applied to an Australian population as in its derivation population. The relative simplicity of the PWH score makes it a viable tool for clinical use, although utility of such tools may be more suited for research in determining inclusion or exclusion criteria for comparative outcome studies.