Background: Traumatic brain injury (TBI) is a disruption to normal brain functions caused by traumas such as collisions, blows, or penetrating injuries. There are factors affecting patient outcomes that also have a predictive value. Limited data from low-middle income countries showed a high number of poor outcomes in TBI patients. The corticosteroid randomisation after significant head injury (CRASH) prognostic model is a predictive model that uses such factors and is often used in developed countries. The model has an excellent discriminative ability. However, there is still a lack of studies on its use in surgical patients in low-middle income countries. This study aimed to evaluate the CRASH model's validity to predict 14-day mortality of TBI patients who underwent surgery in low-middle income countries.
Methods: This retrospective analytical observational study employed total sampling including all TBI patients who underwent surgery with general anesthesia from January to December 2022. Statistical analysis was performed by applying Mann-Whitney and Fisher exact tests, while the model's discriminative ability was determined through the area under the curve (AUC) calculations.
Results: 112 TBI patients were admitted during the study period, and 74 patients were included. Independent statistical analysis showed that 14-day mortality risk, age, Glasgow Coma Scale score, TBI severity, pupillary response, and major extracranial trauma had a significant individual correlation with patients' actual mortality outcome (p < 0.05). The AUC analysis revealed an excellent mortality prediction (AUC 0.838; CI 95%).
Conclusion: The CRASH prognostic model performs well in predicting the 14-day mortality of TBI patients who underwent surgery in low-middle income countries.