The objective of this study was to verify if replacing the Injury Severity Score (ISS) with the New Injury Severity Score (NISS) in the original Trauma and Injury Severity Score (TRISS) form would improve the survival rate estimation. This retrospective study was performed in a level I trauma center for one year. ROC curve was used to identify the best indicator (TRISS or NTRISS) for survival probability prediction. Participants were 533 victims, with a mean age of 38±16 years. There was a predominance of motor vehicle accidents (61.9%). External injuries were more frequent (63.0%), followed by head/neck injuries (55.5%). The survival rate was 76.9%. There is a predominance of ISS scores ranging from 9-15 (40.0%), and NISS scores ranging from 16-24 (25.5%). Survival probability equal to or greater than 75.0% was obtained for 83.4% of the victims according to TRISS, and for 78.4% according to NTRISS. The new version (NTRISS) is better than TRISS for survival prediction in trauma patients ¹⁾.