Comparison of APACHE II, Adjusted APACHE II and TRISS Scores in Predicting Mortality Rate in Head Trauma Patients Admitted to ICU at Khatam-al-Anbia Hospital of Zahedan

Background: Head trauma is a common and potentially fatal injury. Various models such as APACHE II, simplified APACHE (SAPA II), MPM II, and TRISS have been proposed to predict mortality in these patients. In this study, we compared the efficacy of APACHE II, adjusted APACHE II and TRISS models in predicting the mortality rate and quality of provided services in head trauma patients. Materials and Methods: In this retrospective cross sectional study, we calculated the APACHE II and TRISS scores, and predicted mortality rate by each model as well as predicted rate of adjusted APACHE II in head trauma patients admitted to a general ICU at Khatam-al-Anbia hospital of Zahedan from June 2006 to June 2007. The discriminating power of the models to predict the mortality rates were then measured and compared using area under ROC curve and z statistics, respectively. P value less than 0.05 was considered statistically significant. Results: Fifty six patients were included in this cross-sectional study. Of these, there were 49 men and 7 women with the mean age of 24.4 years. The mean duration of ICU admission was 10.07±6.40 days and 11 patients (19.6%) passed away. The mean APACHE II and TRISS scores were 44.78±5.13 and 18.75±8.79, respectively. The comparison between the APACHE II and adjusted APACHE II (P value = 0.8), APACHE II and TRISS (P value=0.2), and adjusted APACHE II and TRISS (P value=0.25) showed no significant difference. Discussion: Our study revealed that APACHE II predicted the mortality rates better than adjusted APACHE II and TRISS models in head trauma patients admitted to ICU but this priority was not statistically significant.