As the leading cause of mortality and disability in trauma patients, traumatic brain injury (TBI) brings damage to victims and their families' quality of life and economic burden to society. A recent study estimated that nearly 69 million individuals are diagnosed with TBI annually in the world (1). The high mortality of TBI patients makes early triage and clinical intervention extremely important. Many trauma scores have been developed to assess the injury severity of TBI and to predict the outcome of trauma patients, such as the revised trauma score (RTS), the injury severity score (ISS), and the comprehensive trauma revised injury severity score (TRISS) (2–5). Specific scoring systems aimed at predicting the outcome of TBI patients were also developed and validated, including the CRASH model and IMPACT model (6, 7). However, these scores composed of radiologic characteristics and other anatomical factors are too complex and time-consuming for clinicians making patient triage and treatment decisions in the early stage after initial brain injury.

The BIG (composed of base deficit, international normalized ratio, and Glasgow Coma Scale) score is a pediatric trauma score that was initially developed to assess children facing military and civilian traumatic injuries (8). It has been proven to accurately predict the mortality rate of pediatric trauma patients admitted to military trauma systems. Several subsequent studies externally confirmed the good performance of the BIG score in predicting the mortality of similar pediatric trauma patients (9–11). Moreover, one study confirmed that the BIG score in admission was associated with functional outcomes at hospital discharge in pediatric TBI patients (12). The BIG score performed better than other pediatric trauma scoring systems and was validated with similar accuracy in a separate pediatric population (8). A BIG score of <12 points suggests a mortality of <5%, whereas a cutoff of >26 points corresponds to a mortality of >50% (13). In addition, researchers also explored the prognostic value of the BIG score in non-specific adult trauma patients and found that the BIG score had a comparable predictive performance with TRISS and the probability of survival (PS09) score (14). Given that aging is a factor that affects the prognosis of trauma patients, we proposed to establish an age-adjusted BIG score to better predict the mortality of patients with trauma.

It was mentioned in all the above studies that the superiority of the BIG score was its availability and simplicity. Based on these findings, we designed this study to explore the prognostic value of the BIG score and compared it with other trauma triage scores in homogeneous adult TBI patients.

In our study, the BIG score was not a valuable risk stratification tool for adult TBI patients. However, the age-adjusted BIG score performed well in predicting outcomes with an AUC of 0.764. The prognostic value of the age-adjusted BIG score was superior to the readily available physiological scoring system RTS. Combining four indicators of mortality (age, base deficit, international normalized ratio, and Glasgow Coma Scale) in trauma patients, the age-adjusted BIG score could comprehensively reflect the injury severity and possible progression of adult traumatic patients.

As an indicator of shock, base deficit was confirmed to be associated with injury severity and mortality in pediatric and adult trauma patients (17–20). Researchers also concluded that base deficit was a useful predictor of coagulation decompensation and shock-related complications after trauma (21, 22). One study found that an increased base deficit was associated with prolonged partial thromboplastin and prothrombin times and low protein C levels in trauma patients (23). A Evaluating level of base deficit could help physicians decide the requirements of early transfusion to avoid the development of hypoperfusion (21, 22). Maintaining appropriate blood pressure and cerebral perfusion was necessary to alleviate secondary brain injury in TBI patients. A previous study including pediatric TBI patients illustrated that patients with poor outcomes had a higher base deficit level than those with good outcomes (12). Our results were similar to the findings of a previous study that indicated that non-survivors had a higher base deficit level than survivors. Another study showed that the base deficit on admission was statistically negatively correlated with GCS and RTS in adult TBI patients (24). However, the level of base deficit did not significantly differ between survivors and non-survivors in this study. This contradictory and unconvincing result might be attributable to the small sample size of this study. Although statistically non-significant after adjusting confounders, the base deficit was significantly related to in-hospital mortality in the univariate logistic regression analysis of our study.

It was estimated that a quarter of patients with severe trauma would present an abnormal blood coagulation test on admission, which would be positively associated with poor outcomes for these patients (25, 26). As a reflection of coagulation function, INR was confirmed as an accurate predictor of mortality and organ failure in trauma patients (27, 28). Acute traumatic coagulopathy (ATC) was primarily caused by the endothelial activation of the protein C pathway, which was induced by tissue injury and hypoperfusion (29, 30). Other factors such as resuscitation-induced hemodilution, hypothermia, and acidosis would aggravate the ATC (26, 31). The prevalence of coagulopathy after TBI had been reported as ranging from 7 to 63%. The huge discrepancy might be attributable to the different definitions of coagulopathy and heterogeneous populations (32–35). Specifically, the TBI itself was independently correlated with the development of coagulopathy due to the excessive fibrinolysis caused by extensive tissue factor release from the injured brain (36, 37). Coagulopathy was acknowledged as an independent risk factor of mortality and neurological outcomes in TBI patients (38–41). Poor coagulation function could increase the potential of intracranial hemorrhage, extracranial hemorrhage, and secondary neuronal loss (26, 36). Some studies indicated that incorporating results of the coagulation test including INR could improve the value of the conventional TBI prognostic model (42, 43). Therefore, the BIG score incorporating INR could reflect the severity and possible progression of TBI patients more comprehensively.

GCS, which has been widely used for nearly five decades, is an indicator of brain injury severity and cerebral perfusion. However, the classic GCS could not be accurately evaluated in intubated, sedated, and intoxicated patients (44, 45). The comparison of the AUC value between the GCS score alone and our age-adjusted BIG score showed that incorporating base deficit, INR, and age into GCS could improve the prognostic value and stability of clinical use in TBI patients. The multivariable prognostic model we constructed was composed of seven factors, namely, age, SpO₂, glucose, hemoglobin, BIG score, subarachnoid hemorrhage, and intracerebral hematoma. Although this model had a significantly higher AUC value than the age-adjusted BIG score, its evaluation was much more complex than the age-adjusted BIG score, which makes it more applicable in hospitalization but not in the emergency department. Instead, the simplicity and easy availability of the age-adjusted BIG score allow it to be quickly evaluated without the consideration of additional factors. This advantage is significant for physicians carrying out patient triage and providing intensive medical therapy for potentially high-risk TBI patients in the early stage after injury. Therefore, the age-adjusted BIG score has been specially applied by emergency department workers.

There were several limitations in this study. First, most of the included patients were those who received treatment in the ICU of a single medical center. Patients with mild TBI might not be included in this study. Nearly half of TBI patients in the database who did not meet inclusion criteria were excluded. Therefore, selection bias could not be avoided. The exact predictive value of the age-adjusted BIG score and the constructed prognostic model should be externally verified by a prospective study in other medical centers. Second, the predictive value of the age-adjusted BIG score was not specifically analyzed in subgroups of included TBI patients, such as patients with a penetrating injury or blunt injury. A previous study showed that the BIG score was more valuable in predicting outcomes of penetrating trauma patients than blunt trauma patients (14). Third, the level of base deficit and INR in admission could be influenced by pre-hospital intubation and resuscitation. Records of these two variables were not collected in the present study. Finally, in addition to RTS, other complex trauma scores such as ISS and TRISS were not evaluated in the present study. A study comparing the predictive value between age-adjusted BIG scores and these scores is worthwhile to conduct in the future. Despite these limitations, the readily available age-adjusted BIG score is more efficient than other complex scores in patient triage and treatment decisions in the early stage after brain injury.

As a pediatric trauma score, the BIG score is not applicable to adult TBI patients. However, the age-adjusted BIG score is a readily available and effective score that is beneficial for clinicians to triage adult TBI patients and evaluate possible progression in the early stage after injury. The prognostic model incorporating the BIG score has a better predictive value and could be used in TBI patients during hospitalization.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

The studies involving humans were approved by the Ethics Committee of the West China Hospital of Sichuan University (NO: 2021-1684). The studies were conducted in accordance with the local legislation and institutional requirements. The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants' legal guardians/next of kin because this observational study was performed using data from the Multiparameter Intelligent Monitoring in Intensive Care Database III (MIMIC-III database). All data of participants in this public database were deidentified and anonymized.

XB: Conceptualization, Data curation, Methodology, Resources, Software, Validation, Writing—original draft, Writing—review & editing. RW: Conceptualization, Data curation, Methodology, Resources, Software, Validation, Writing—original draft, Writing—review & editing. CZ: Investigation, Methodology, Project administration, Writing—original draft. DW: Formal analysis, Methodology, Resources, Supervision, Validation, Writing—review & editing. LM: Data curation, Formal analysis, Methodology, Resources, Supervision, Visualization, Writing—review & editing. MH: Data curation, Formal analysis, Methodology, Supervision, Validation, Visualization, Writing—review & editing.