BACKGROUND: Trauma is a public health problem of vast proportions. It is the leading cause of death among young people. The Major Trauma Outcome Study (MTOS) is a retrospective descriptive study of injury severity and outcome, considered the largest database of descriptive contemporary injuries information. The objective of this paper is to compare the retrospective calculation of NISS with prospectively calculated ISS, using TRISS and a simple modification called NTRISS (New Trauma and Injury Severity Score), and to compare this population submitted to laparotomy with MTOS patients. METHODS: We studied 1.380 trauma adult patients submitted to laparotomy at Discipline of Trauma Surgery - Unicamp, in Campinas, during a 8-year period. Submitted data included: demographics, cause of injury (blunt or penetrating, gunshot or stab wounds), physiologic status at admission (RTS), anatomic injury diagnoses (ATI, ISS and NISS), survival probability using the TRISS and NTRISS, and patient outcomes (survival or death). Z statistic and W statistic, first described by Flora, were used to compare the predicted number of deaths (or survivors) with the baseline MTOS norm. RESULTS: The majority of these patients was male (88.3%) with a mean age of 30.4 years. Gunshot wounds were the most frequent (641 cases - 46.4%) mechanism of trauma. Thirty one per cent had blunt injuries. The median ATI, ISS and NISS values were, respectively, 12.3, 17.6 and 22.1. The overall mortality rate was 16.8% and the patients with blunt trauma had the highest mortality rate (29.3%). NISS better separated survivors from nonsurvivors than ISS, with higher specificity of NTRISS. It was observed a significantly less survivors than expected from outcomes norms (Z -16.24 with TRISS and Z -9.40 with NTRISS). Ranges of W values for each patient set demonstrated a difference in the numbers of nonsurvivors of 7.89 more deaths per 100 patients treated than expected from the MTOS using TRISS, while this values were reduced to 5.14 using NTRISS. CONCLUSION: The methods used for survival probability calculation presented limitations, particularly in this population with predominance of penetrating trauma. The NISS, with its derived NTRISS, was the score that better predicted survival than ISS. The results obtained with TRISS and NTRISS were significantly worse than MTOS, but this process of monitoring trauma patients have been important to ensure continued provision of quality of care.
Trauma; Trauma scoring; Laparotomy; TRISS
