D. Stewart1, C. Janowak1, H. Jung1, A. Liepert1, A. O’Rourke1, S. Agarwal1 1University Of Wisconsin,Surgery,Madison, WI, USA
Introduction: Many methods exist for predicting mortality among adult trauma patients; however, most systems ignore patient co-morbidity, a significant predictor of outcome, in their calculations. The American Society of Anesthesiologists Physical Status (ASA-PS), a well-validated and easy-to-use scale, is an assessment of pre-operative status that has been shown to accurately predict post-operative mortality. Using the ASA-PS as a marker of cumulative patient comorbidity severity we sought to test whether we would be able to improve the predictive power of the Injury Severity Score (ISS), the most commonly utilized trauma grading system, with respect to mortality, major complication, and discharge disposition.
Methods: A retrospective review of a prospectively collected and internally validated database at an academic Level I trauma center was performed for consecutive adult admissions between 2009-2013. Abbreviated Injury Scale (AIS) was measured by region (head/neck, face, thorax, abdomen, extremities, general) and severity of injury (1 to 5). ISS was measured by summing the squares of the three most injured regions [(AIS1)2 + (AIS2)2 + (AIS3)2]. ASA-PS scores were assigned based on patient comorbidities and then integrated with the traditional ISS in a variety of permutations, including adjustments of ASA-PS for patient age >70 and using individual AIS components of ISS. We assessed these various models for predictive ability with a primary outcome of mortality and secondary outcomes of major complications as per National Trauma Data Bank (NTDB) definitions as well as discharge disposition using receiver operating characteristic (ROC) analysis. These were compared with the ISS.
Results: All of the ISS/ASA-PS hybrid formulas outperformed ISS alone in predictive power for mortality, major complication, and discharge disposition. The best overall permutation, (AIS1)2+(AIS2)2+(Age-Modified ASA-PS)2, yielded an ROC of 0.888 for mortality as compared to ISS with an ROC=0.853 (p<0.001). Similar differences were seen for discharge disposition (Hybrid ROC=0.743; ISS ROC=0.639, p<0.001) and major complication (Hybrid ROC=0.761; ISS ROC=0.719, p<0.001).
Conclusion: Incorporating ASA-PS into calculations of trauma scoring is both simple and more predictive of mortality, major complication, and discharge disposition than the traditional ISS metric. Replacing ISS with this new method, which takes patient age and comorbid condition into account through adaptation of the ASA-PS improves prognostication of outcomes and enables care providers to prioritize resources for injured patients.
