Introduction: measuring the severity of traumatic injuries is crucial for predicting clinical outcomes. Whereas the Injury Severity Score (ISS) has limitations in assigning scores to injuries at the same site, the New Injury Severity Score (NISS) corrects for this problem by taking into account the three most severe injuries regardless of the region of the body. This study seeks to comprehend the clinical and epidemiological profile of trauma patients while comparing the effectiveness of scales for predicting mortality. Methods: a descriptive, observational and retrospective study using records of patients who underwent thoracotomy at the Hospital das Clínicas of the Federal University of Triângulo Mineiro between 2000 and 2019. Demographic data, mechanisms of injury, affected organs, length of stay and mortality were analyzed. Injury severity was assessed using the ISS and NISS, and statistical analyses were conducted using MedCalc and SigmaPlot. Results: 101 patients were assessed, on average 29.6 years old, 86.13% of whom were men. The average duration of hospitalization was 10.9 days and the mortality rate was 28.7%. The ROC curve analysis revealed a sensitivity of 68.97%, specificity of 80.56% and area under the curve of 0.837 for the ISS, and 58.62%, 94.44% and 0.855 for the NISS, respectively. The Youden index was 0.49 for the ISS and 0.53 for the NISS. Conclusion: the study demonstrated comparable efficacy of NISS and ISS in predicting mortality. These findings hold significance in the hospital setting. Professionals must be familiar with these scales to utilize them competently for each patient.