Abstract
Background: Traumatic brain injury (TBI) is a major cause of morbidity and mortality in older adults. While age and severity are known predictors of outcomes, limited research exists on how specific TBI types and sizes affect prognosis in older age groups.
Methods: A retrospective cohort study analyzed data from the American College of Surgeons Trauma Quality Programs Participant Use File (ACS-TQIP-PUF) (2017–2022) including 182,661 adults ≥40 years with isolated blunt TBI (subdural hematoma >8mm, epidural hematoma >8mm, contusion >2cm, subarachnoid hemorrhage, or diffuse axonal injury [DAI]) and no skull fractures. Polytrauma cases were excluded. Outcomes included in-hospital mortality, intensive care unit (ICU) length of stay (LOS), and ventilator days. Multivariable logistic and linear regression models were used, adjusting for demographic, clinical, and injury characteristics.
Results: Elderly patients (≥65 years) showed significantly higher mortality across all TBI types, most notably with contusions >2cm (OR 4.85, 95% CI 3.46–6.80, P<.01). DAI in elderly patients was associated with the longest ICU LOS (+2.63 days, 95% CI 1.76–3.49, P<.01) and ventilator duration (+4.29 days, 95% CI 1.90–6.67, P<.01). Factors such as anticoagulation use, low Glasgow Coma Scores, and specific comorbidities further increased risk.
Conclusions: Older adults with larger, isolated TBIs, especially contusions and DAI, experience worse clinical outcomes. These findings underscore the critical need for age-tailored trauma management protocols and early intervention strategies to reduce mortality and improve recovery in this vulnerable population.
- Diffuse Axonal Injury
- Elderly Patients
- ICU Outcomes
- Mortality Risk
- Subdural Hematoma
- Traumatic Brain Injury
Traumatic brain injury (TBI) is a major public health concern and a leading cause of death and disability worldwide. In the United States alone, TBIs contribute to approximately 30% of all injury-related deaths annually, with older adults experiencing disproportionately worse outcomes following head trauma.1 As the population continues to age, the burden of TBI among individuals aged 65 years and older is increasing rapidly, both in incidence and in clinical complexity. Geriatric patients often present with diminished physiological reserves, comorbidities, and altered responses to injury and treatment, all of which complicate the management of TBI and heighten the risk of adverse outcomes.2 While prior research has established age as an independent risk factor for increased morbidity and mortality following TBI, most studies have focused on generalized measures of injury severity, such as the Glasgow Coma Scale (GCS) or the Abbreviated Injury Scale (AIS), without examining the nuanced impact of specific intracranial hemorrhage types or lesion sizes.3,4 Additionally, many investigations have grouped all TBIs together, failing to account for the considerable heterogeneity in injury patterns, such as subdural hematoma (SDH), epidural hematoma (EDH), cerebral contusion, subarachnoid hemorrhage (SAH), and diffuse axonal injury (DAI), each of which may carry distinct prognostic implications. This limitation reduces the clinical utility of such studies, as treatment decisions often depend on the specific type and extent of injury visible on neuroimaging.
Emerging literature has attempted to stratify outcomes by TBI subtype, yet few studies have explored the role of lesion size, such as SDH thickness or contusion diameter, in predicting outcomes, especially in elderly populations.5 Studies that do address this question tend to be limited by small sample sizes, single-institution data, or a focus on operative management rather than broader clinical outcomes. Moreover, previous research often neglects the influence of common confounding factors such as comorbid conditions, anticoagulation use, and initial presentation characteristics, all of which are more prevalent and impactful in older adults.6 The knowledge gap is further compounded by a lack of comparative data between middle-aged adults (40–64 years) and the elderly (≥65 years), particularly in relation to outcomes such as intensive care unit (ICU) length of stay (LOS), ventilator dependence, and early mortality. Understanding how these outcomes vary across different age groups and injury types is critical to improving triage protocols, guiding surgical decisions, and optimizing resource allocation in trauma care systems. Notably, the current literature does not clearly define whether certain TBI subtypes pose disproportionately higher risks for older adults, nor does it adequately quantify the effect of age on intensive care needs and ventilator days across distinct hemorrhage patterns.
Given these limitations, this study aimed to fill the existing knowledge gap by leveraging a large national trauma dataset to investigate the association between TBI type and size with key clinical outcomes, mortality, ICU LOS, and ventilator days, in middle-aged and elderly adults with isolated blunt TBIs. By controlling for a wide array of demographic, clinical, and injury-related variables, this study seeks to provide more granular insights into age-related disparities in TBI outcomes. Such findings may inform age-specific care guidelines and promote more tailored approaches to managing traumatic brain injuries in older populations.
Objective
The primary objective of this study was to evaluate the impact of TBI type and lesion size on early clinical outcomes, specifically in-hospital mortality, ICU LOS, and ventilator days, among middle-aged (40–64 years) and elderly (≥65 years) patients with isolated blunt TBI. Recognizing the increased vulnerability of older adults to TBI-related complications, this study aimed to determine whether specific intracranial injury patterns, such as SDH >8mm, EDH >8mm, contusion >2cm, SAH, and DAI, are independently associated with worse outcomes in elderly populations compared to their middle-aged counterparts. By leveraging a large, nationally representative trauma data set and adjusting for critical demographic, clinical, and injury-related variables, the study sought to fill existing knowledge gaps and generate evidence-based insights that can inform risk stratification, management strategies, and resource allocation in the care of older TBI patients.
Methods
Study design and data source
This investigation employed a retrospective cohort design using data from the American College of Surgeons Trauma Quality Programs Participant Use File (ACS-TQIP-PUF) for the years 2017 through 2022. Institutional Review Board oversight was waived under exemption [RHO10224], in accordance with applicable regulatory standards. The study adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines, with corresponding documentation available in Supplementary Table 1 (available online). The ACS-TQIP-PUF is a de-identified, nationwide dataset derived from over 700 participating trauma centers in the United States, ranging from Level I to Level V facilities, including centers with unspecified designations.7 All included centers submit standardized trauma data to the National Trauma Data Bank (NTDB), ensuring broad representation and uniform data quality.
For this analysis, data were retrospectively reviewed to examine associations between specific variables and clinical outcomes in patients with isolated blunt TBI. Variables of interest included TBI characteristics (type and lesion size), patient demographics (age, sex, race, ethnicity, insurance status, trauma center verification level, and transport method), and injury metrics such as Injury Severity Score (ISS) and Abbreviated Injury Scale (AIS) scores. Clinical outcomes assessed were in-hospital mortality, ICU LOS, and ventilator days. Procedural data (intracranial pressure [ICP] monitoring, craniotomy) identified by ICD-10 codes and procedure timing, were also included. The administration of venous thromboembolism prophylaxis (low molecular weight heparin [LMWH] and unfractionated heparin [UFH]) was reported. The presence of key comorbid conditions, including anticoagulation or bleeding disorders, was incorporated into the analysis to account for clinical complexity.
Study population and eligibility criteria
The study population included adult patients aged 40 years and older who sustained an isolated TBI (Figure 1). Individuals were excluded if they had a skull fracture. This exclusion was applied to reduce confounding, as skull fractures often represent a distinct injury mechanism and are associated with higher rates of complications such as cerebrospinal fluid leaks, cranial nerve injuries, and open fractures, which may require unique surgical interventions and carry a different prognosis. By excluding these cases, the study ensured a more homogenous population for evaluating outcomes specifically related to intracranial hemorrhage and parenchymal brain injuries.
Participant Flowchart Among Middle Aged to Elderly Adults (≥ 40 Years) Who Had a Larger Isolated Blunt Traumatic Brain Injury (Larger Isolated SDH, EDH, Contusion, or DAI, SAH) Without a Skull Fracture. Note: ACS-TQIP=American College of Surgeons Trauma Quality Improvement; TBI=traumatic brain injury; ISS=injury severity score; AIS=abbreviated injury severity score; SDH=subdural hematoma; EDH=epidural hematoma; SAH=subarachnoid hemorrhage
Isolated TBI was defined by injury type and lesion size using AIS codes, which categorize injury characteristics based on computed tomography (CT) imaging performed within the first 24 hours of admission (Supplementary Table 2, available online).8 To further isolate brain-specific injuries, patients were excluded if they had an AIS severity score >2 in non-cranial regions categorized as ISS Regions 2 through 6 (face, chest, abdomen or pelvis, extremities or pelvic girdle, and external). This exclusion criterion ensured extracranial injuries did not contribute to the observed outcomes, allowing for a focused assessment of TBI-specific morbidity and mortality. Patients with small-volume injuries, specifically SDH and EDH ≤8 mm and contusions ≤2 cm, were excluded from the analysis to focus on larger injuries with clearer clinical relevance. The final study cohort included patients with SDH >8 mm, EDH >8 mm, contusions >2 cm, SAH, and DAI, all of which are known to carry increased risk for adverse outcomes.
The Modified Berne-Norwood Criteria were applied to define thresholds for TBI lesion size and guide group stratification. This classification system is widely used in clinical research and neurosurgical literature to delineate small from large hemorrhagic lesions based on size cutoffs shown to predict clinical deterioration and the need for surgical intervention.9 For example, a SDH >8 mm in thickness or a midline shift over 5 mm has been associated with significantly higher mortality and neurologic decline. Similarly, contusions larger than 2 cm often correlate with increased intracranial pressure and worse neurological outcomes. These thresholds provide an evidence-based framework for identifying TBIs likely to influence outcomes and require higher levels of care. Participants were stratified into two age-based cohorts: middle-aged (40–64 years) and elderly (≥65 years). A total of ten TBI groups were formed according to type and size, enabling robust comparative analysis. The use of the Modified Berne-Norwood Criteria in this context enhances the clinical relevance of the study, allowing researchers to differentiate injuries with significant prognostic implications and support age-specific risk stratification in trauma populations.
Outcomes of interest
The primary outcomes evaluated in this study were in-hospital mortality, ICU LOS, and duration of mechanical ventilation (ventilator days). These outcomes were selected due to their clinical significance in assessing early hospital-based morbidity and mortality following TBI. In-hospital mortality was defined as death occurring at any point during the index hospitalization. ICU LOS was measured in days from ICU admission to ICU discharge, capturing the intensity and duration of critical care required for each patient. Ventilator days represented the total number of days a patient required invasive mechanical ventilation, serving as a surrogate marker for respiratory compromise and neurological impairment. These outcomes were analyzed to explore differences in severity and resource utilization among various TBI types and sizes, with a particular focus on age-related disparities between middle-aged and elderly patients. By examining these early hospital outcomes, the study aimed to identify injury patterns associated with increased clinical risk and to inform triage, treatment, and discharge planning in TBI care pathways.
Statistical analysis
Descriptive statistics were used to summarize the characteristics of the study population, including means, medians, interquartile ranges for continuous variables, and frequencies with percentages for categorical variables. Group comparisons for continuous variables were performed using the Kruskal-Wallis test, while categorical variables were compared using Pearson’s Chi-square test of independence. All analyses were conducted using R version 4.4.1 (R Core Team, 2023). Prior to regression modeling, multicollinearity diagnostics were performed to assess predictor suitability. Variables demonstrating tolerance values >0.5 were retained to ensure model stability and meet the assumptions of independence and goodness-of-fit. All retained variables were determined to be statistically independent and appropriately powered for inclusion.
To evaluate the associations between TBI type and size with key outcomes, both logistic and linear regression models were constructed. Logistic regression was used to assess the primary binary outcome of in-hospital mortality, while linear regression models were employed to evaluate continuous outcomes, including ICU LOS and duration of mechanical ventilation. These models were developed using SPSS version 28 (IBM Corp., Armonk, NY). The main exposure of interest in all models was TBI type and lesion size, stratified by age group (40–64 years vs. ≥65 years). Covariates included in the adjusted models were selected based on clinical relevance and prior literature, and included patient age, sex, race, ethnicity, mode of transportation, trauma center verification level, GCS score, type of venous thromboembolism prophylaxis, neurosurgical interventions (ICP monitoring, craniotomy), and pre-existing comorbid conditions. A P value of <0.05 was considered statistically significant for all analyses.
Results
Section I: first analysis patient description
A total of 182,661 adult patients aged ≥40 years with isolated blunt TBI were included in the final analysis. The most common injury subtype was isolated SAH, accounting for 73.6% of the cohort (n = 134,546), followed by SDH >8 mm (25.0%, n = 45,756), DAI (0.4%, n = 845), EDH >8 mm (0.5%, n = 931), and cerebral contusion >2 cm (0.3%, n = 583). Among elderly patients (≥65 years) with isolated TBI, those with large cerebral contusions (>2 cm; n = 387) and DAI (n = 258) constituted two of the smallest but clinically significant subgroups. The mean age of patients in the contusion group was 78.4 years (SD = 6.9), compared to 74.7 years SD = 6.9) in the DAI group (P < .01). Male sex predominated in both groups—59.9% in the contusion group and 67.4% in the DAI group (P < .01). Caucasians comprised the majority of both cohorts, representing 80.8% of those with contusions and 79.8% with DAI (P < .01).
Comorbidities were more frequent in the contusion group, with hypertension (67.4%), diabetes (22.2%), dementia (16.5%), and anticoagulant use (32.0%) commonly observed. DAI patients also had a notable comorbidity burden, including hypertension (51.9%) and anticoagulation (15.8%), though overall rates were lower than in the contusion group (all P < .01). Neurologic severity at presentation was significantly different between groups. The median Glasgow Coma Scale (GCS) score was 14 (IQR=7) in contusion patients, indicating a mild to moderate injury, whereas DAI patients had a median GCS of 7 (IQR=12), consistent with severe neurologic dysfunction and coma (P < .01). Similarly, AIS head scores, which rate anatomical severity from 1 (minor) to 6 (unsurvivable), were significantly higher in the DAI group (mean AIS 4.4; SD=.4) than in the contusion group (mean AIS 4.1; SD=.3; P < .01), suggesting greater intracranial injury burden. Falls were the predominant mechanism of injury in both groups but occurred more frequently in the contusion group (92.5%) than in the DAI group (58.5%; P < .01). In terms of acute care utilization, the mean ICU LOS was 8.5 days (SD=9.3) for DAI compared to 4.3 days (SD=4.3) for contusions (P < .01), and ventilator dependence followed a similar trend. The most striking outcome was in-hospital mortality, which reached 35.2% in elderly DAI patients versus 27.6% in those with contusions (P < .01). These findings emphasize the disproportionately severe trajectory of DAI and in the elderly, characterized by deeper coma at presentation, higher anatomic injury scores, greater ICU needs, and significantly elevated risk of death.
Section II: second analysis, regression models
Logistic regression associated with in-hospital mortality
Multivariable logistic regression analysis identified several factors significantly associated with increased in-hospital mortality among patients aged ≥40 years with large isolated blunt TBI, including age-stratified injury types, clinical variables, neurosurgical interventions, and comorbidities (Table 1). Across all TBI subtypes, advanced age (≥65 years) was independently associated with significantly higher odds of mortality when compared to patients aged 40–64 years. The mortality risk was particularly elevated in the elderly with contusions >2 cm (odds ratio [OR] 4.85, 95% confidence interval [CI] 3.46–6.80, P < .01), followed by EDH >8 mm (OR 2.88, 95% CI 2.05–4.05, P < .01), DAI (OR 2.74, 95% CI 1.76–4.28, P < .01), and SDH >8 mm (OR 2.80, 95% CI 2.59–3.03, P < .01). In contrast, elderly patients with SAH had only a modest increase in mortality risk (OR 1.10, 95% CI 1.02–1.20, P = .01).
Logistic regression analysis of factors associated with in-hospital mortality among middle aged to elderly adults (≥ 40 years) who had larger isolated blunt traumatic brain injury (larger isolated SDH, EDH, contusion or SAH, DAI) without a skull fracture
Among clinical covariates, a lower GCS score was strongly associated with higher mortality (OR 0.74 per point increase, 95% CI 0.74–0.74, P < .01), emphasizing the predictive power of neurologic status at admission. Being treated at a Level I trauma center was also associated with increased mortality risk (OR 1.20, 95% CI 1.13–1.26, P < .01), which may reflect referral bias and greater injury severity in these settings. Arrival via ground ambulance was linked to slightly higher odds of mortality (OR 1.13, 95% CI 1.06–1.21, P < .01), potentially indicating delayed or less aggressive prehospital intervention.
Neurosurgical interventions had a mixed impact. The presence of an ICP monitor was associated with significantly increased mortality (OR 1.76, 95% CI 1.60–1.94, P < .01), likely reflecting the severity of injuries warranting monitoring. Undergoing craniotomy was also linked to higher mortality (OR 1.10, 95% CI 1.02–1.18, P = .01), although the effect size was smaller. Conversely, LMWH prophylaxis was protective, associated with a significantly reduced risk of in-hospital death (OR 0.62, 95% CI 0.60–0.65, P < .01), suggesting that appropriately timed venous thromboembolism prophylaxis may play a role in improving outcomes in select TBI patients. Several comorbidities were also significant predictors. The strongest risk factors included the presence of an advanced directive (OR 2.24, 95% CI 2.07–2.43, P < .01), renal failure (OR 2.20, 95% CI 1.97–2.46, P < .01), congestive heart failure (CHF) (OR 1.56, 95% CI 1.43–1.69, P < .01), and anticoagulation use (OR 1.52, 95% CI 1.43–1.61, P < .01). Additional significant predictors included alcohol use disorder (OR 1.17, P < .01), chronic obstructive pulmonary disease (COPD) (OR 1.31, P < .01), and diabetes (OR 1.09, P < .01). Interestingly, mental health diagnoses were associated with reduced mortality risk (OR 0.82, 95% CI 0.76–0.89, P < .01), warranting further exploration into possible psychosocial or healthcare access factors.
Linear regression associated with ICU LOS
ICU LOS varied significantly by TBI subtype and age group (Table 2). Compared to patients aged 40–64 years, elderly patients (≥65 years) with DAI experienced a prolonged ICU stay, with an average increase of +2.63 days (95% CI: 1.76 to 3.49, P < .01). In contrast, elderly patients with other TBI types—SDH >8 mm, EDH >8 mm, contusions >2 cm, and SAH—had shorter ICU stays than their younger counterparts. For example, elderly patients with SDH >8 mm had a reduction of −0.51 days (95% CI: −0.62 to −0.40, P < .01), and those with EDH >8 mm showed a −0.72-day difference (95% CI: −1.28 to −0.16, P = .01). These findings suggest that while elderly patients with DAI require more intensive monitoring, those with other hemorrhagic injuries may receive shorter ICU care, potentially reflecting early withdrawal of care or less aggressive management.
Linear regression analysis of factors associated with intensive care unit length of stay among middle aged to elderly adults (≥ 40 years) who had larger isolated blunt traumatic brain injury (larger isolated SDH, EDH, contusion or SAH, DAI) without a skull fracture
Higher GCS scores were independently associated with shorter ICU stays (β = −0.24 per point increase; 95% CI: −0.25 to −0.23, P < .01), reinforcing the role of initial neurologic function in predicting care intensity. Conversely, the use of LMWH for venous thromboembolism prophylaxis was associated with a +1.41-day increase in ICU stay (95% CI: 1.36 to 1.45, P < .01), likely due to cautious monitoring following anticoagulation in high-risk patients. Neurosurgical procedures were among the strongest predictors of prolonged ICU utilization. The presence of an ICP monitor was associated with an increase of +3.97 ICU days (95% CI: 3.78 to 4.17, P < .01), while craniotomy contributed an additional +3.08 ICU days (95% CI: 2.95 to 3.21, P < .01), reflecting the expected intensity of care for patients requiring surgical or intracranial pressure management.
Linear regression associated with ventilation days
The duration of mechanical ventilation varied notably by injury type and age (Table 3). Elderly patients (≥65 years) with DAI required significantly longer ventilatory support compared to their younger counterparts, with an average increase of +4.29 days (95% CI: 1.90 to 6.67, P < .01). As expected, more severe neurological impairment was associated with prolonged ventilation. Patients who required ICP monitoring experienced a substantial increase in ventilator days (+5.68 days, 95% CI: 5.14 to 6.22, P < .01), while those who underwent craniotomy had an average increase of +3.26 days (95% CI: 2.91 to 3.62, P < .01). These findings emphasize the intensity of care associated with surgical management of severe TBI. Administration of low molecular weight heparin (LMWH) for venous thromboembolism (VTE) prophylaxis was independently associated with longer ventilator time (+2.37 days, 95% CI: 2.24 to 2.49, P < .01), likely due to delayed initiation in critically ill patients and the need for cautious monitoring.
Linear regression analysis of factors associated with ventilation days among middle aged to elderly adults (≥ 40 years) who had larger isolated blunt traumatic brain injury (larger isolated SDH, EDH, contusion or SAH, DAI) without a skull fracture
Discussion
In this large national cohort of adults aged 40 years and older with isolated blunt TBI, we identified substantial age-related differences in outcomes based on injury type and size. The most striking findings were observed among elderly patients (≥65 years) with DAI and large cerebral contusions (>2 cm). These individuals experienced the highest in-hospital mortality, longest ICU stays, and most prolonged mechanical ventilation durations, despite representing only a small proportion of the total sample. In contrast, although SAH was the most prevalent injury subtype, it was associated with lower mortality and reduced resource utilization relative to DAI and large contusions. Notably, lower GCS scores, presence of ICP monitoring, and comorbid conditions such as renal failure and CHF were independently associated with poorer outcomes, consistent across regression models. These findings carry significant implications for acute trauma care and critical care triage. In particular, the disproportionately poor outcomes among elderly patients with DAI and contusions suggest a need for early aggressive monitoring, goal-directed management, and perhaps enhanced prognostication tools specific to geriatric TBI subtypes. The extended ICU and ventilator durations in these groups place additional demands on healthcare systems and emphasize the importance of aligning treatment goals with anticipated outcomes. Additionally, the protective association observed with the use of LMWH in relation to mortality supports the judicious use of venous thromboembolism prophylaxis even in this high-risk population, echoing findings from other large-scale trauma analyses.10
Our results align with and expand upon existing literature showing that elderly patients fare worse following TBI, particularly when presenting with low GCS or requiring neurosurgical interventions.11,12 However, few prior studies have stratified outcomes by specific injury subtypes and lesion sizes. By using the Modified Berne-Norwood Criteria, our study provides a more granular assessment of TBI severity and highlights the underappreciated lethality of large contusions and DAI in older adults. While past studies have emphasized the prognostic significance of midline shift or hematoma thickness, our study extends these insights by showing how age modifies the relationship between injury type and clinical trajectory.13 Interestingly, while patients with large SDH (>8 mm) also had elevated mortality, the effect size was less pronounced than for contusions and DAI in the elderly subgroup. This may reflect differences in clinical management, such as the more frequent use of surgical decompression in SDH cases or the variable impact of anticoagulation, both of which warrant further exploration. Future investigations should focus on developing age- and injury-specific prognostic models that integrate imaging findings, functional status, and comorbid profiles. Prospective studies are needed to validate whether early intervention strategies, such as expedited neurosurgical consultation, goal-concordant care discussions, and early venous thromboembolism prophylaxis can improve outcomes in high-risk elderly TBI subgroups. Additionally, qualitative research exploring decision-making processes around ICU admission and surgical intervention in the elderly may provide valuable context for our findings. Finally, integrating frailty indices and pre-injury cognitive status into predictive models could enhance clinical risk stratification beyond age and injury metrics alone.
Limitations
This study has several limitations that should be considered when interpreting the findings. First, the analysis was based on retrospective data from the ACS-TQIP database, which, while comprehensive and nationally representative, is inherently subject to limitations related to coding accuracy, missing data, and variability in reporting practices across institutions. Although standardized data collection methods are employed by TQIP, potential misclassification of TBI type, lesion size, or comorbid conditions remains a possibility. Additionally, the use of AIS codes to classify hemorrhage type and size relies on initial computed tomography interpretations, which may not capture interval changes or the dynamic progression of brain injuries over time.
Second, although we applied strict exclusion criteria to isolate patients with blunt TBI without skull fractures or significant extracranial trauma, residual confounding by unmeasured clinical factors, such as frailty, pre-existing cognitive impairment, or goals-of-care decisions may influence the observed associations, particularly among elderly patients. For example, shorter ICU stays or reduced surgical intervention rates in older adults may reflect treatment limitations rather than differences in injury severity.
Third, causality cannot be established due to the observational study design. While we identified strong associations between TBI type, age, and outcomes, it is not possible to determine whether these injuries directly caused increased mortality or prolonged ICU and ventilator use, particularly in the context of complex comorbidities and care decisions. Additionally, outcomes such as long-term neurologic recovery, functional status, and quality of life were not captured in this dataset, limiting our ability to evaluate the full impact of TBI beyond the acute care setting.
Lastly, while the use of large national data enhances generalizability, it may not reflect practice patterns in non-TQIP centers or international settings. Variation in neurosurgical decision-making, ICU protocols, and thresholds for initiating palliative care could influence outcomes and may limit the applicability of findings across all trauma systems. Despite these limitations, this study leverages a large, rigorously collected dataset and provides valuable insight into the age-specific impact of TBI type and size on clinical outcomes. The findings underscore the need for future prospective studies that incorporate functional outcomes, neuroimaging, and patient-centered goals of care to further clarify risk and guide management strategies in older adults with traumatic brain injury.
Conclusion
In this large, national cohort study of adults aged 40 years and older with isolated blunt traumatic brain injury, we found injury type and size, when stratified by age, have significant implications for in-hospital outcomes. Elderly patients with large cerebral contusions and DAI experienced the highest rates of mortality, longest ICU stays, and most prolonged ventilator dependence, highlighting these subgroups as particularly vulnerable. In contrast, SAH, though most common, was associated with comparatively lower clinical burden. These findings emphasize the importance of early risk stratification and age-specific care strategies, particularly in older adults with high-risk TBI subtypes. The results also support the integration of injury-specific classifications, such as the Modified Berne-Norwood Criteria, into trauma assessment tools to improve outcome prediction and guide clinical decision-making. Moving forward, tailored interventions, early discussions of goals of care, and prospective validation of these findings will be essential to optimize outcomes and resource allocation for aging trauma populations with complex intracranial injuries.
Acknowledgment
TQP PUF Admission 2017 to 2022 Version (e.g., 2020.1.0), Chicago, IL, 20XX. The content reproduced from the TQP PUF remains the full and exclusive copyrighted property of the American College of Surgeons. The American College of Surgeons is not responsible for any claims arising from works based on the original data, text, tables, or figures.
Footnotes
Funding: This project was funded, in part, through philanthropic support of Marshfield Clinic Research Institute led by the Marshfield Clinic Health System Foundation (255800-00-RES SUPPT TRAUMA).
Disclosures: The authors have stated there are no conflicts of interest related to this work.
- Received June 10, 2025.
- Revision received December 1, 2025.
- Revision received January 8, 2026.
- Accepted January 19, 2026.
References
- 1.↵Taylor CA, Bell JM, Breiding MJ, Xu L. Traumatic Brain Injury-Related Emergency Department Visits, Hospitalizations, and Deaths - United States, 2007 and 2013. MMWR Surveill Summ. 2017;66(9):1-16. doi:10.15585/mmwr.ss6609a1
- 2↵Alvis BD, Hughes CG. Physiology Considerations in Geriatric Patients. Anesthesiol Clin. 2015;33(3):447-456. doi:10.1016/j.anclin.2015.05.003
- 3.↵Miller GF, Daugherty J, Waltzman D, Sarmiento K. Predictors of traumatic brain injury morbidity and mortality: Examination of data from the national trauma data bank: Predictors of TBI morbidity & mortality. Injury. 2021;52(5):1138-1144. doi:10.1016/j.injury.2021.01.042
- 4.↵Liu C, Xie J, Xiao X, Clinical predictors of prognosis in patients with traumatic brain injury combined with extracranial trauma. Int J Med Sci. 2021;18(7):1639-1647. doi:10.7150/ijms.54913
- 5.↵Podolsky-Gondim GG, Cardoso R, Zucoloto Junior EL, . Traumatic Brain Injury in the Elderly: Clinical Features, Prognostic Factors, and Outcomes of 133 Consecutive Surgical Patients. Cureus. 2021;13(2):e13587. Published 2021 Feb 27. doi:10.7759/cureus.13587
- 6.↵Ghneim M, Brasel K, Vesselinov R, Traumatic Brain Injury in Older Adults: Characteristics, Outcomes, and Considerations. Results From the American Association for the Surgery of Trauma Geriatric Traumatic Brain Injury (GERI-TBI) Multicenter Trial. J Am Med Dir Assoc. 2022;23(4):568-575.e1. doi:10.1016/j.jamda.2022.01.085
- 7.↵American College of Surgeons. Trauma quality programs participant use file. Trauma Quality Programs Participant Use File. Available at: https://www.facs.org/quality-programs/trauma/quality/national-trauma-data-bank/datasets/
- 8.↵Association for the Advancement of Automotive Medicine. (2018). Abbreviated Injury Scale: 2015 Revision. 6 ed. Chicago, IL.
- 9.↵Pastorek RA, Cripps MW, Bernstein IH, The Parkland Protocol’s modified Berne-Norwood criteria predict two tiers of risk for traumatic brain injury progression. J Neurotrauma. 2014;31(20):1737-1743. doi:10.1089/neu.2014.3366
- 10.↵Jacobs BN, Cain-Nielsen AH, Jakubus JL, . Unfractionated heparin versus low-molecular-weight heparin for venous thromboembolism prophylaxis in trauma. J Trauma Acute Care Surg. 2017;83(1):151-158. doi:10.1097/TA.0000000000001494
- 11.↵Kehoe A, Smith JE, Bouamra O, Edwards A, Yates D, Lecky F. Older patients with traumatic brain injury present with a higher GCS score than younger patients for a given severity of injury. Emerg Med J. 2016;33(6):381-385. doi:10.1136/emermed-2015-205180
- 12.↵Ostermann RC, Joestl J, Tiefenboeck TM, Lang N, Platzer P, Hofbauer M. Risk factors predicting prognosis and outcome of elderly patients with isolated traumatic brain injury. J Orthop Surg Res. 2018;13(1):277. doi:10.1186/s13018-018-0975-y
- 13.↵de Souza MR, Fagundes CF, Solla DJF, . Mismatch between midline shift and hematoma thickness as a prognostic factor of mortality in patients sustaining acute subdural hematoma. Trauma Surg Acute Care Open. 2021;6(1):e000707. doi:10.1136/tsaco-2021-000707
