Predicting mortality in traumatic intracranial hemorrhage

Area of Science:

• Neurosurgery
• Trauma Surgery
• Critical Care Medicine

Background:

• Traumatic intracranial hemorrhage (tICH) is a major cause of death and disability in trauma patients.
• Existing prognostic models for tICH have limitations in accuracy, generalizability, and practicality.
• There is a need for a simpler, more accurate risk stratification method for tICH patients.

Purpose of the Study:

• To develop a simpler and more accurate risk stratification model for all traumatic intracranial hemorrhage patients.
• To improve the identification of high-risk tICH patients.
• To aid in clinical decision-making and resource allocation for tICH management.

Main Methods:

• Retrospective analysis of 3564 tICH patients from a level 1 trauma center (2003-2013).
• Collected data included demographics, vital signs, GCS, ISS, and hemorrhage type.
• K-fold cross-validation and Akaike Information Criterion were used to develop and select the best risk stratification models.

Main Results:

• The developed model demonstrated good calibration (p < 0.001) and high concordance (0.881).
• Factors associated with increased mortality included older age, lower blood alcohol concentration, antiplatelet/anticoagulant use, lower GCS, and higher ISS.
• Successful risk stratification was achieved across various hemorrhage subtypes.

Conclusions:

• A novel, accurate, and well-calibrated risk stratification model for tICH was developed.
• The model is applicable to multiple hemorrhage subtypes and can assist in resource allocation and clinical decision-making.
• Future research will focus on external validation and the creation of a simplified, mentally calculable version of the model.