Evaluation of mortality prediction using SOFA and APACHE IV tools in trauma and non-trauma patients admitted to the ICU

Background Various tools have previously been introduced to predict the recuperation and mortality of patients in intensive care units and to classify them, which have particular advantages and disadvantages compared to each other. The present study compared the prediction power of mortality of trauma and non-trauma patients admitted to the ICU by SOFA and APACHE IV tools. Methods In this retrospective cohort study, patients admitted to the ICU of Kowsar Hospital in Sanandaj from the beginning of April 2020 to the end of December 2020 were assessed. Data were collected in the form of a questionnaire based on APACHE IV and SOFA criteria as well as the demographic information questionnaire. All collected data related to the first 24 h of patients' hospitalization was analyzed in SPSS V16 software using Chi-square, Mann-Whitney, Cox regression and Pearson correlation coefficient. Results This study was performed on 404 patients admitted to the ICU, Out of which 273 people (67.6%) were male, 208 (51.5%) trauma patients and 196 (48.5%) non-trauma ones. Patients' mean age was 54.76 +/- 20.77 years and their average length of stay in the hospital was 10.05 +/- 8.49 days. In general, the AUC obtained by APACHE IV tool (0.902) was slightly better than that of SOFA tool (0.895). However, in a specific study of traumatic and non-traumatic patients, it was found that APACHE IV and SOFA tools had better performance in predicting death innon-trauma and trauma patients based on the accuracy, AUC and sensitivity, respectively. Conclusions Based on the results of this study, the difference between APACHE IV and SOFA tools in predicting death of patients admitted to the ICU was very small but the function of APACHE IV was better in predicting mortality of non-traumatic patients, while the function of SOFA was better in predicting the death of traumatic cases. This represents the applicability of these two tools in different patient subgroups.