ABSTRACT
OBJECTIVE: The goal of this study was to compare the prediction performance of two anatomic scales, the Injury Severity Scale (ISS) and the New Injury Severity Scale (NISS), with two physiologic scales, the Revised Trauma Scale (RTS) and the Simplified Acute Physiology Scale II (SAPS II), in trauma patients. DESIGN: Prospective study carried out over a 16-month period. SETTING: Emergency department of a teaching hospital. PATIENTS: Hospitalized victims of trauma up to 14 years of age. INTERVENTIONS: The primary endpoint was the survival status at hospital discharge; the secondary outcome was need for ICU admission. Model discrimination was evaluated by the area under the receiver-operating characteristic curve and model calibration was evaluated using the Hosmer-Lemeshow goodness-of-fit statistic. MEASUREMENTS AND MAIN RESULTS: A total of 1136 patients, with an average age of 37.6 years, fulfilled the inclusion criteria. The mortality rate was 4.5%. The combined rate of hospital death and ICU admission was 17.3%. The ISS and the NISS showed excellent discriminative power for mortality prediction (AUC 0.94 and 0.93, respectively) and ICU admission decision (0.91 and 0.89, respectively), and a good calibration. The SAPS II and the RTS showed lower discriminative power. Combining ISS or NISS with SAPS II did not improve significantly the predictive performance of each scale alone. CONCLUSION: Both ISS and NISS showed better predictive severity performance compared with RTS and SAPS II in trauma patients. The combination of anatomic scales with physiologic ones did not improve the prediction performance of each scale considered alone.
