Enhancing Mortality Probability Model II (MPM II) Predictive Accuracy with the Lethal Triad in ICU Trauma Patients

Abstract

The Mortality Probability Model II (MPM II) is a well-recognized predictive tool in the intensive care unit (ICU) scoring system. The presence of a lethal triad, including hypothermia, acidosis, and coagulopathy, is associated with a worse outcome for trauma patients. This study aimed to assess whether integrating the lethal triad could enhance the MPM II predictive accuracy for trauma patients in the ICU. The study conducted a retrospective analysis of adult trauma patients admitted to an ICU from January 1, 2016, to December 31, 2022, in a Level I trauma center in southern Taiwan. It assessed the impact of the lethal triad variables incorporated into MPM II scores using the area under the receiver operating characteristic curve (AUC of ROC). Of the 3,410 patients included, 257 had the lethal triad and 3,153 do not. The mortality rate was significantly higher in patients presenting the lethal triad than those without (46.7% vs. 8%, p < 0.001). However, there was no significant improvement in predictive accuracy with both the MPM II with or without incorporating the variable of lethal triad having an area under the Receiver Operating Characteristic curve of 0.893. While the lethal triad is important for understanding trauma-induced physiological changes, straight inclusion into MPM II has no substantial prognostic value. These findings indicate that the existing variables in the MPM II may already be indirectly capturing the impacts of the lethal triad, emphasizing the intricate interplay of physiological components in trauma patients. This underscores the importance of continued study in improving the prognostic modeling to ultimately predicting outcomes for trauma patients.