Objectives: To determine the usefulness of head Abbreviated Injury Score (AIs), Injury Severity Score (ISS), and Glasgow Coma Scale (GCS) as measures of injury severity and predictors of outcome after traumatic brain injury (TBI). Hypothesis: We hypothesize that AIS and ISS correlate better with outcome following a TBI than GCS, and that a combination of both the anatomic and physiologic measures would be better predictors than either measure alone. Participants: The cohort consisted of 279 subjects with TBI from a Level I trauma center. Brain injuries were confirmed by the presence of abnormal neuroimaging or postresuscitation GCS of less than 13, and the requirement for inpatient hospitalization. Methods: Clinical information was prospectively collected on age, gender, GCS, head AIS, ISS, and computed tomographic results. Data were also collected on injury severity, initial computed tomographic scan, length of intensive care unit stay, length of hospital stay, and discharge disposition. Outcomes were measured at 12 months postinjury using the GOS-E via telephone, mailed questionnaires, or face-to-face interviews. Results: ISS was the best predictor of GOS-E (rs = -0.316, P < .001), followed by Head AIS (rs = -0.212, P < .001) and GCS (rs = 0.190, P < .001). When considered in combination, GCS and ISS modestly improved the correlation with GOS-E (R = 0.274, P < .001). The combination of GCS and Head AIS had a similar effect (R = 0.229, P = .001). Correlations were stronger from patients younger than 48. Conclusions: Anatomic measures modestly outperform GCS as predictors of GOS-E. The combination of GCS and AIS/ISS correlates with outcome better than any of the three measures alone. Results support the addition of anatomic measures in clinical studies of TBI. In addition, most of the variance in outcome is not accounted for by currently available measures of injury severity.

REFERENCES