INTRODUCTION
Traumatic brain injury (TBI) is a leading cause of morbidity and mortality worldwide, significantly affecting both individuals and healthcare systems. Globally, the incidence of TBI continues to rise, with millions of new patients each year contributing to a substantial burden on emergency departments (EDs) and long-term care facilities [1–4]. TBI is often associated with considerable long-term disability, making it a critical focus for clinicians and researchers aiming to improve patient outcomes [5,6].
TBI rarely occurs in isolation, with many patients receiving additional injuries to other anatomical regions that complicate both diagnosis and management [1,6]. These concomitant injuries can exacerbate the patient’s condition, leading to more complex treatment protocols and variable prognoses [7,8]. The distinction between isolated TBI (iTBI) and non-isolated TBI (niTBI) is clinically important: whereas iTBI refers to patients whose brain injury is the only significant trauma, niTBI involves concurrent injuries to other parts of the body [9].
Despite the growing body of research on TBI, a significant gap remains in understanding the differential outcomes of iTBI and niTBI patients. Previous studies have often failed to make this distinction, treating TBI as a homogeneous entity without adequately considering the influence of concomitant injuries [2,10]. Understanding the differences in outcomes between these groups is essential to improving care strategies and optimizing resource allocation [7,8]. Specifically, there is a lack of comprehensive, multicenter data examining the effects of concomitant injuries on the clinical outcomes of TBI patients.
Our primary objective in this study was to evaluate the effects of concomitant injuries on clinical outcomes in patients with TBI, specifically comparing in-hospital mortality within the first 24 hours and 30 days, hospital admission rates, and interhospital transfers between iTBI and niTBI patients. Our secondary objective was to perform a detailed analysis of how different anatomical sites of concomitant injuries affect clinical outcomes within the niTBI group.
METHODS
Ethics statement
This study’s protocol was reviewed and approved by the Institutional Review Board of Jeju National University Hospital (No. JEJUNUH 2024-06-006). Informed consent was waived due to the use of deidentified data and the study’s retrospective nature.
Study design
This retrospective cross-sectional study used data from a multicenter prospective trauma registry known as the Emergency Department-based Injury In-depth Surveillance (EDIIS), which is operated by the Korean Disease Control and Prevention Agency. The EDIIS database is a robust and comprehensive epidemiological dataset of injuries that has been meticulously collected from 23 EDs in tertiary hospitals across Korea. Data were systematically collected for all patients presenting with injuries at the EDs of these participating hospitals. The data collection process was conducted in real-time, capturing detailed information from the point of the patient's arrival at the ED through to their discharge or transfer, thereby ensuring thorough and continuous tracking of each case.
Data sources
The EDIIS registry, compiled by ED physicians at the participating hospitals, contains comprehensive data on TBI: detailed patient demographics (sex, insurance status, vital signs, and mental status), injury characteristics (intention, mechanism, activities, location, anatomical site of injury, use of emergency medical services [EMS], and alcohol involvement), injury severity, emergency care procedures, diagnosis, treatment, ED disposition (e.g., discharge, interhospital transfer, admission, death), and post-admission outcomes (e.g., discharge, interhospital transfer, death).
In terms of injury severity, the registry records both the Revised Trauma Score (RTS) and the Excess Mortality Ratio-adjusted Injury Severity Score (EMR-ISS). The EMR-ISS is a robust scoring system developed in 2009 to assess injury severity within large-scale patient datasets and is based on the International Classification of Diseases, 10th edition (ICD-10) [11].
Study population
The study population comprised all patients aged 0 to 120 years who presented to the participating EDs with TBI between January 1, 2020, and December 31, 2022. TBI patients within the EDIIS database were defined using specific diagnostic codes from the ICD-10 [12]: S01.0–S01.9, S02.0–S02.1, S02.3, S02.7–S02.9, S04.0, S06.0–S06.9, S07.0–S07.1, S07.8–S07.9, S09.7–S09.9, T01.0–T02.0, T04.0, and T06.0. Those codes were used irrespective of the presence of additional injury diagnoses. Patients declared dead on arrival and those with missing outcome data were excluded to maintain data integrity.
Following those exclusions, the final analyzed population was stratified into two distinct groups: iTBI and niTBI. The iTBI group contained patients who sustained an injury exclusively to the head, as evidenced by diagnostic codes, and the niTBI group comprised patients with concomitant injuries in addition to the TBI.
The classification of patients into the iTBI and niTBI groups was rigorously performed using all the injury diagnostic codes within the EDIIS database. The EDIIS system records up to 10 diagnostic variables per patient, each based on the ICD-10 classification system. The first three characters of each code specify the anatomical location of the injury. For example, codes starting with "S00–S09" indicate injuries to the head, and codes beginning with other prefixes such as "S10–S19" (neck) and "S20–S29" (thorax) correspond to injuries in different anatomical regions. Patients were classified as having iTBI if all the diagnostic variables contained ICD-10 codes beginning with "S00–S09," indicating injuries confined solely to the head region, without any associated injuries to other anatomical areas. Conversely, patients were categorized as having niTBI if any diagnostic variable contained an ICD-10 code with a prefix outside of "S00–S09," signifying the presence of additional injuries in regions other than the head [12].
Statistical analysis
We conducted a comparative analysis of baseline demographics and injury epidemiologic characteristics between the iTBI and niTBI groups. For continuous variables, we assessed normality using the Shapiro-Wilk test. We then used the Student t-test for normally distributed data and the Wilcoxon rank sum test for non-normally distributed data. For categorical variables, we used the chi-square test or Fisher exact test, as appropriate. Descriptive statistics are presented as means with standard deviations for normally distributed continuous variables, medians with interquartile ranges for non-normally distributed continuous variables, and frequencies with percentages for categorical variables.
The primary outcome of interest was in-hospital mortality, specifically mortality within 24 hours and within 30 days of the ED visit. Mortality outcomes were determined based on the recorded time of death in relation to the patient’s initial ED presentation. Although the registry does not contain separate predefined variables specifically denoted as “24-hour mortality” or “30-day mortality,” we calculated these outcomes by comparing the documented time of death with the time of the patient’s arrival in the ED. Patients who died within 24 hours of their ED visit were classified as having 24-hour mortality, and those who died within 30 days were classified as having 30-day mortality. The secondary outcomes were hospital admission and interhospital transfer, reflecting resource utilization in the ED. Univariate logistic regressions were initially performed to analyze the dichotomous primary and secondary outcomes. Variables with a P-value of <0.20 in the univariate analyses were considered for inclusion in the multivariate logistic regression models. Multivariate logistic regressions were then used to adjust for potential confounders: age, sex, insurance type, EMS usage, injury intention, activity at the time of injury, injury location (indoor/outdoor), alcohol involvement, and whether the injury occurred on a weekend. Adjusted odds ratios (aORs) with 95% confidence intervals (CIs) were compared between the iTBI and niTBI groups to assess the influence of concomitant injuries on 24-hour mortality, 30-day mortality, hospital admission, and interhospital transfer.
To further elucidate the effects of specific concomitant injury sites on clinical outcomes within the niTBI group, we conducted a detailed subgroup analysis. Patients in the niTBI group were categorized based on the presence or absence of injuries to specific anatomical regions, as identified by their ICD-10 diagnostic codes: neck (S10–S19), thorax (S20–S29), abdomen, pelvis, and lower back (S30–S39), shoulder and upper arm (S40–S49), elbow and forearm (S50–S59), wrist, hand, and fingers (S60–S69), hip and thigh (S70–S79), knee and lower leg (S80–S89), ankle and foot (S90–S99) [12]. For each anatomical site, we performed multivariate logistic regression analyses to calculate aORs for the primary outcomes (24-hour mortality and 30-day mortality) and secondary outcomes (hospital admission and interhospital transfer). These regression models were adjusted for the same set of potential confounders as the overall analysis.
All statistical analyses were conducted using Stata ver. 18.0 (StataCorp). Statistical significance was set at a two-tailed P-value of less than 0.05.
RESULTS
Patient selection
From January 1, 2020, to December 31, 2022, the EDIIS database recorded 590,734 injury patients. After excluding 410,577 non-TBI patients and 99 TBI patients (2 with undetermined outcomes and 97 dead on arrival), we identified 180,058 TBI patients for analysis. These were categorized into the iTBI group (n=127,673, 70.9%) and niTBI group (n=52,385, 29.1%) (Fig. 1).
Demographic characteristics
Patient characteristics are shown in Table 1. Overall, the mean age of the study population was 36.6±27.6 years. Patients in the niTBI group were significantly older, with a mean age of 46.3±23.7 years, than those in the iTBI group, whose mean age was 32.6±28.1 years (P<0.001). The iTBI group had a higher proportion of women (35.6% vs. 33.5%, P<0.001) and more patients covered by the Korean National Health Insurance (90.6% vs. 71.5%, P<0.001), whereas automobile insurance coverage was more common in the niTBI group (18.3% vs. 3.7%).
Systolic and diastolic blood pressures, pulse rate, and respiration rate all differed significantly between the groups. The niTBI group also had lower mean Glasgow Coma Scale (GCS) scores (14.4±2.2 vs. 14.7±1.4, P<0.001) and a higher proportion of severe TBI (GCS, 3–8) patients (4.3% vs. 1.8%).
The injury severity scores indicated more severe injuries in the niTBI group, with lower RTS (7.7±0.6 vs. 7.8±0.5, P<0.001) and higher EMR-ISS (26.5±15.5 vs. 12.1±9.1, P<0.001). Major trauma was more frequent among niTBI patients (9.3% vs. 6.3%, P<0.001), as was the need for surgical intervention (11.2% vs. 3.2%, P<0.001).
The niTBI group had a longer median ED length of stay (3.4 hours vs. 2.2 hours, P<0.001) and different ED dispositions: fewer were discharged directly from the ED (70.6% vs. 89.9%), and more required hospital admission (24.5% vs. 8.4%), interhospital transfer (3.6% vs. 1.1%), or died in the ED (1.1% vs. 0.3%; all, P<0.001).
Injury epidemiologic characteristics
Significant differences in injury characteristics were observed between groups (Table 2). Intentional injuries were more frequent in the niTBI group (9.6% vs. 4.4%, P<0.001). Slips and falls were more common in iTBI patients (52.4% vs. 44.5%), and motor vehicle collisions (MVCs) were significantly more prevalent in niTBI patients (34.3% vs. 9.3%, P<0.001). Compared with iTBI patients, niTBI patients were more often injured during leisure or play (25.2% vs. 20.3%) and less often during vital activities (42.6% vs. 59.5%, P<0.001). Road-related injuries were more frequent in the niTBI group (49.0% vs. 23.3%), whereas indoor injuries were more common in iTBI patients (59.1% vs. 31.6%, P<0.001).
EMS usage was higher among niTBI patients (51.0% vs. 25.9%, P<0.001), and alcohol involvement was more common (19.4% vs. 13.4%, P<0.001). There was no significant difference in injuries occurring during daytime hours, but slightly more weekend injuries occurred in the iTBI group (36.7% vs. 34.5%, P<0.001).
Clinical outcomes between the iTBI and niTBI groups
The niTBI patients had higher 24-hour mortality (1.5% vs. 0.4%), 30-day mortality (2.6% vs. 1.0%), hospital admission rates (24.5% vs. 8.4%), and interhospital transfers (3.6% vs. 1.1%) than iTBI patients (all P<0.001) (Table 3).
Table 4 presents the results of univariate and multivariate logistic regression analyses assessing the effects of concomitant injuries on the primary outcomes (mortality within 24 hours and 30 days) and secondary outcomes (hospital admission and interhospital transfer) among patients with TBI. The multivariate logistic regression analyses show that, after adjusting for confounders, niTBI patients had higher odds of 24-hour mortality (aOR, 1.456; 95% CI, 1.286–1.648; P<0.001) and 30-day mortality (aOR, 1.111; 95% CI, 1.022–1.208; P=0.014). They were also more likely to be admitted to the hospital (aOR, 2.175; 95% CI, 2.107–2.245; P<0.001) and require interhospital transfer (aOR, 2.016; 95% CI, 1.868–2.177; P<0.001).
Anatomical injury sites and their effects in the niTBI group
Table 5 details the distribution of injury sites among patients in the niTBI group, providing insight into the relationship between specific injury locations and clinical outcomes. Table 6 presents the results of multivariate logistic regression analyses evaluating the association between injuries to specific anatomical sites and clinical outcomes (24-hour mortality, 30-day mortality, hospital admission, and interhospital transfer).
Within the niTBI group, a higher number of concomitant injury sites correlated with increased mortality: patients who died within 24 or 30 days had a median of three injury sites, compared with two in those admitted or transferred.
Thoracic injuries were strongly associated with adverse outcomes; they were present in 67.6% of 24-hour mortalities (517 of 765) and increased the odds of 24-hour mortality nearly sixfold (aOR, 5.958; 95% CI, 5.057–7.019; P<0.001). These injuries also elevated the odds of 30-day mortality, hospital admission, and interhospital transfer. Abdominal, pelvic, and lower back injuries significantly increased the odds of 24-hour mortality (aOR, 2.741; 95% CI, 2.343–3.206; P<0.001) and 30-day mortality, as well as hospital admission and transfer rates. Hip and thigh injuries were associated with higher odds of 24-hour mortality (aOR, 2.703; 95% CI, 2.217–3.294; P<0.001) and hospital admission (Fig. 2).
Conversely, neck injuries, although present in patients with adverse outcomes, were less predictive of mortality (aOR, 0.436; 95% CI, 0.350–0.544; P<0.001). Injuries to the elbow and forearm and wrist, hand, and fingers were associated with lower odds of mortality but higher odds of interhospital transfer (Fig. 2).
DISCUSSION
In this study, we explored the effects of concomitant injuries on clinical outcomes in patients with TBI by comparing iTBI and niTBI patients. We found that niTBI patients had significantly worse outcomes: higher mortality rates, increased hospital admissions, and more frequent interhospital transfers. They also differed from iTBI patients in demographics, injury mechanisms, and patterns. These findings contribute to an understanding of how concomitant injuries influence TBI outcomes, filling a gap in the literature and offering insights for improved management and policy development in trauma care [7–9,13–16].
The niTBI group, comprising 29.1% of the study population, presented with more severe clinical conditions, evidenced by lower GCS scores and higher injury severity scores. The higher incidence of MVCs and work-related injuries in this group indicates that high-energy trauma contributes to sustaining multiple injuries. The niTBI patients were more often injured during paid work or leisure activities and on roads, with greater EMS usage and a higher prevalence of alcohol-related injuries, highlighting additional risk factors.
Our observations align with prior studies indicating that MVCs are a leading cause of niTBI [2,7,9,15]. The predominance of male patients in the niTBI group reflects known epidemiological patterns, as men are more involved in high-risk activities and occupations [1,9,13,14]. A systematic review highlighted sex differences in TBI incidence and outcomes, noting that men are disproportionately affected due to factors such as occupational hazards and engagement in risky behaviors [10]. According to the Centers for Disease Control and Prevention, MVCs are one of the primary causes of TBI, especially in younger populations [2]. The high-energy impact associated with MVCs often results in multiple injuries, including severe TBI and thoracic and abdominal trauma. The higher prevalence of alcohol-related injuries among niTBI patients corresponds with evidence that alcohol consumption increases the risk of sustaining multiple severe injuries due to impaired judgment and risk-taking behaviors [17–19].
Our results confirm that niTBI patients have significantly worse clinical outcomes than iTBI patients, with higher 24-hour mortality (1.5% vs. 0.4%) and 30-day mortality (2.6% vs. 1.0%). Our multivariate analyses show that concomitant injuries independently increased the odds of 24-hour mortality (aOR, 1.456) and 30-day mortality (aOR, 1.111) after adjusting for potential confounders, suggesting that they contribute to the overall trauma burden. This is supported by previous studies indicating that extracranial injuries, particularly thoracic and abdominal injuries, are associated with higher mortality rates in TBI patients [7–9,13–16]. Those studies suggested that systemic inflammatory responses and physiological stress induced by extracranial injuries could contribute to secondary brain injury, worsening neurological outcomes [15]. These findings highlight the need for integrated management of brain and extracranial injuries, emphasizing early detection and treatment to improve survival, consistent with prior research [13–15].
The niTBI patients required more hospital admissions (24.5% vs. 8.4%) and interhospital transfers (3.6% vs. 1.1%) than the iTBI patients, with aORs of 2.175 and 2.016, respectively. This increased resource utilization aligns with other studies showing that patients with multiple injuries have longer hospital stays, more intensive care unit admissions, and higher healthcare costs than those with a single injury, emphasizing the need for efficient resource allocation and specialized trauma systems [9,15].
Within the niTBI group, certain injury sites were strongly associated with adverse outcomes. Thoracic injuries were the most significant predictor, increasing the odds of 24-hour mortality nearly sixfold (aOR, 5.958) and the odds of hospital admission more than threefold (aOR, 3.329). Injuries to the abdomen, pelvis, lower back, hip, and thigh were also significantly associated with higher mortality and hospital admissions.
Conversely, injuries to the neck, elbow and forearm, and wrist, hand, and finger were associated with decreased mortality odds. This suggests that not all concomitant injuries affect outcomes equally; lower energy injuries such as falls from standing height can result in less systemic stress than higher-energy injuries, such as those from MVCs. Previous studies have noted better survival rates in patients with isolated limb injuries than in those with central or multiple severe injuries [7,15,20]. Recognizing these associations can aid in triaging and resource allocation, with early intervention for high-risk injury patterns to potentially improve survival and reduce transfers.
Our findings highlight the need for comprehensive trauma assessment and management protocols for TBI patients with concomitant injuries. Early recognition and management of thoracic and abdominal injuries are crucial, as recommended by the American College of Surgeons Committee on Trauma [5]. EDs should prioritize the rapid identification of thoracic and abdominal injuries, which are significant predictors of increased mortality and hospital admission [1,6,21]. Targeted prevention strategies are also necessary. The higher incidence of MVCs and work-related injuries among niTBI patients suggests that improving road safety measures and enforcing occupational safety protocols could reduce severe multisystem injuries.
Limitations
This study has several limitations. First, its retrospective design and reliance on a registry database could introduce selection bias and limit the ability to control for all confounding variables. Second, we included patients of all ages (0–120 years), but the trauma patterns and outcomes in very young and very old patients might differ markedly from those in other age groups due to distinct physiologic responses and comorbidity profiles. We included age as a covariate in our analyses to adjust for these differences, but future research could stratify patients into narrower age groups or consider separate pediatric and geriatric cohorts for more nuanced insights.
Third, we did not incorporate standardized injury severity scales (e.g., Abbreviated Injury Scale or Injury Severity Score) or account for preexisting conditions, in-hospital complications, or institutional care variations in our regression models. Although our primary objective was to determine how the presence and anatomical locations of concomitant injuries influence clinical outcomes in TBI patients, we recognize that differences in injury severity between the iTBI and niTBI groups might have influenced the observed results. For instance, the niTBI group demonstrated significantly lower RTS and higher EMR-ISS, suggesting an overall greater severity of injury. In this analysis, we considered injury severity as a post-injury factor closely tied to outcome variables, and thus we focused our adjustments on demographic and epidemiologic characteristics that precede or coincide with the injury event. By not directly incorporating severity measures into the multivariate models, we sought to isolate the effect of concomitant injuries rather than potentially conflating their impact with underlying severity. Nonetheless, future research would benefit from the inclusion of standardized severity scoring systems to more clearly distinguish the influence of injury severity from the effects of concomitant injuries.
Fourth, the study period overlapped with the COVID-19 pandemic, which might have affected trauma patterns due to lockdowns and reduced activities, potentially limiting the generalizability of our findings to nonpandemic periods [22,23]. Although the relationship between multiple injuries and clinical outcomes might initially seem like a reaffirmation of existing knowledge, it takes on added significance within the context of our study’s unique methodology and timing. The strength of this work lies in its extensive, multicenter dataset, encompassing 180,058 TBI patients, and data collection during the COVID-19 pandemic. This approach offers robust statistical power and ensures a comprehensive perspective on injury patterns and outcomes. Future studies should consider comparing data from pre- and post-pandemic periods or using analytical methods that adjust for pandemic-related influences on trauma care and patient outcomes.
Conclusions
The presence of concomitant injuries in TBI patients is associated with significantly worse clinical outcomes and increased healthcare resource utilization. Thoracic and abdominal injuries, in particular, substantially elevate the risk of adverse outcomes. These findings underscore the importance of prompt recognition and management of concomitant injuries and highlight the need for targeted prevention strategies.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation 
Print 
