Abstract
Objective: To identify factors influencing neurological prognosis following traumatic brain injury (TBI) and to analyze the role of brain tissue oxygen pressure (PbtO2) monitoring in prognostication. Methods: In this case-control study, medical records of 412 individuals diagnosed with TBI were thoroughly examined and analyzed. The patients were divided into two groups based on their prognosis at three months post-injury: Good Prognosis (n = 321) and Poor Prognosis (n = 91). Demographic and clinical characteristics, brain tissue oxygen partial pressure, radiological and laboratory findings, treatment interventions, and complications were compared between the two groups. Logistic regression analysis was conducted to identify the risk factors for neurological prognosis, and the predictive value of these factors was evaluated using receiver operating characteristic (ROC) curve analysis. Results: The study identified associations between Injury Severity Score (ISS), Glasgow Coma Scale (GCS), PbtO2 levels, radiological findings (diffuse axonal injury and subarachnoid hemorrhage), and laboratory parameters (platelet count and arterial oxygen partial pressure (PaO2)) with neurological prognosis following TBI. Initial PbtO2 levels demonstrated independent predictive value for poor neurological outcomes (Area Under the Curve (AUC) = 0.804). Conclusion: The study highlights the prognostic significance of injury severity, brain tissue oxygenation, radiological findings, and laboratory parameters in determining neurological outcomes following TBI. Furthermore, the findings emphasize the potential of PbtO2 monitoring as a valuable tool in prognostic assessment.
Keywords: Traumatic brain injury, neurological prognosis, influence factors, brain tissue oxygen pressure
Introduction
Traumatic brain injury (TBI) is a significant public health concern due to its pervasive impact on individuals across all age groups and its substantial socioeconomic burden. TBI encompasses a broad spectrum of injuries resulting from external forces, such as blunt trauma or penetrating head injuries, often resulting from motor vehicle accidents, falls, sports-related incidents, and assaults [1,2]. Globally, TBI affects millions of individuals each year, imposing a considerable burden on healthcare systems and society [3,4]. TBI is particularly concerning due to its association with significant morbidity and mortality, exerting a substantial toll on affected individuals and their families [5-7]. While TBI affects individuals across the lifespan, it is notably prevalent among young individuals, with peak incidence observed in adolescents as well as young adults. The long-term consequences of TBI can be profound, leading to a range of physical, emotional, cognitive, and behavioral impairments that can profoundly impact an individual’s quality of life and functional independence [8,9].
Over the past few years, considerable progress has been made in understanding the pathological mechanisms and prognostic factors associated with TBI [10]. Factors such as the Injury Severity Score (ISS), Glasgow Coma Scale (GCS), and radiological findings like diffuse axonal injury and subarachnoid hemorrhage have been consistently identified as critical determinants of neurological outcomes post-TBI [11]. These elements reflect the extent of primary and secondary brain injuries, influencing recovery trajectories and long-term functional outcomes [12]. Among these, physiological monitoring parameters have gained particular interest, specifically brain tissue oxygen partial pressure (PbtO2) monitoring [13]. PbtO2, a measure of cerebral oxygenation, offers real-time insights into the brain’s metabolic state and serves as a sensitive marker of cerebral hypoxia-a known factor exacerbating secondary brain injury [13]. Its role in prognostication is increasingly recognized, as it potentially bridges the gap between the global and regional assessments of brain oxygenation, providing a more nuanced understanding of cerebral pathology in TBI patients [14]. Despite these insights, integrating PbtO2 monitoring into routine clinical practice requires robust evidence of its prognostic value, highlighting the necessity to systematically investigate its utility [15].
The multifaceted nature of TBI pathophysiology necessitates a comprehensive approach to prognostication and management, considering the diverse factors involved in the intricate interplay of primary and secondary injury mechanisms [16-18]. This underscores the need for a multidimensional understanding of TBI pathophysiology to inform the development of tailored therapeutic strategies and prognostic models aimed at improving patient outcomes. Despite advances in acute management strategies, predicting neurological prognosis following TBI remains challenging, necessitating a comprehensive understanding of the diverse factors that influence patient outcomes. This study aims to elucidate the influencing factors of neurological prognosis after TBI and to perform an analysis role of PbtO2 monitoring in this context.
Materials and methods
Study design
This retrospective case-control study analyzed data from 412 TBI patients admitted to Baotou Central Hospital from January 2023 to December 2023. The patients were divided into two groups based on their prognosis three months post-injury: Good prognosis group (n = 321) and Poor prognosis group (n = 91). This study was approved by the Ethics Committee of Baotou Central Hospital and informed consent was waived.
Inclusion and exclusion criteria
Inclusion criteria [19,20]: 1. Definitive diagnosis of acute TBI based on medical history and auxiliary examinations; 2. GCS score of 3-14; 3. Follow-up duration of more than three months. Exclusion criteria: 1. Age under 14 years; 2. Pregnant women; 3. History of autoimmune disease or gout.
The inclusion and exclusion criteria were carefully selected to ensure a homogeneity of the study population and strengthen the validity of the findings. Patients under 14 years of age were excluded due to the physiological and anatomical differences in pediatric populations compared to adults, which could impact the generalizability of brain oxygenation data. Children and adolescents exhibit distinct patterns of injury, recovery, and neurodevelopment that differ significantly from adults, potentially introducing variability that could obscure the analysis of PbtO2’s prognostic value. Additionally, individuals with a history of autoimmune disease or gout were excluded to minimize factors such as chronic inflammation and metabolic disturbances. Autoimmune diseases and gout can independently affect systemic oxygenation and cerebral metabolism, complicating the interpretation of PbtO2 measurements as purely reflective of traumatic injury effects. By excluding these patients, the study aims to focus more precisely on the pathophysiology and prognostic outcomes of TBI in a controlled adult population.
Grouping method
The Glasgow Outcome Score (GOS) was used to evaluate the prognosis of the two patient groups at 3 months post-injury, with 1 point for death, 2 for vegetative state, 3 for severe disability, 4 for moderate disability, and 5 for good recovery. A GOS score > 3 is defined as a favorable prognosis. Patients were grouped based on their prognosis score, with GOS scores > 3 categorized into the Good Prognosis group, and patients with GOS scores ≤ 3 categorized into the Poor Prognosis group.
Clinical characteristics
ISS scores (injury severity)
The severity of injuries across nine body regions (face, head, neck, abdomen, thorax, spine, lower extremities, upper extremities, and external) was assessed using the Abbreviated Injury Scale (AIS), which ranges from 0 to 51. The scale classifies injuries by severity, where 0 for no injury, 1 - a minor injury, 2 - a moderate injury, 3 - a serious injury, 4 - a severe injury, and 5 - a critical injury.
To compute the ISS, these nine body areas were consolidated into six broader categories: R1 for head or neck injuries, R2 for face, R3 for chest, R4 for abdominal or pelvic parts, R5 for extremities or the pelvic girdle, and R6 for external injuries. The ISS was determined by summing the squares of the AIS scores for the three most severe injuries, producing a scale range of 0 to 75. The ISS demonstrated a high reliability, with a Cronbach’s alpha of 0.93 [21].
Admission glasgow coma scale (GCS) score
The GCS is a clinical tool used to assess the level of consciousness in patients by evaluating three aspects: eye response, verbal response, and motor response. The scores for each component are summed, with higher scores indicating better consciousness levels. The maximum score is 15, indicating clear consciousness. Scores of 13-15 indicate mild consciousness impairment, 9-12 indicate moderate impairment, and below 8 indicate coma. Within the scale, a score of 13-15 was categorized as mild, 9-12 as moderate, and 3-8 as severe. The GCS demonstrated a Cronbach’s alpha of 0.78 [22].
Brain tissue oxygen partial pressure detection
Three months post-injury, PbtO2 was measured using the LICOX-II PbtO2 monitoring system (GMS, Germany) with a Clark-type microelectrode probe. The PbtO2 monitoring probe was placed at the junction of the necrotic brain tissue (hematoma) and the penumbra of contused brain tissue. The levels of PbtO2 (mmHg) were recorded, with PbtO2 < 15/20/25/30 mmHg (%) representing the percentage of time at each measurement.
Radiological findings
Three months post-injury, patients underwent brain Computed Tomography (CT) examinations using the Philips Brilliance 64-slice/128-layer spiral CT scanner. The scan parameters included a slice thickness and reconstruction interval of 0.45 mm, a scanning speed of 0.75 r/s, a tube voltage of 120 kV, and a tube current of 250 mA, applying an adaptive iterative dose reduction algorithm. The scan range extended from the inferior margin of the second cervical vertebra to the outer table of the skull. Via a median cubital vein, iodixanol (320 mgI/mL) of a dose of 50-60 mL and 30 mL of normal saline was injected at a flow rate of 4.5-5.5 mL/s. Arterial-phase images were acquired at 2-second intervals, starting 10 seconds post-injection, to capture cross-sectional views. For suspected aneurysms in the circle of Willis, an enlarged scan technique was employed whenever possible. The original images were transferred to the Philips IntelliSpace Portal workstation for post-processing, including volume rendering, multiplanar reformation, curved planar reformation, and maximum intensity projection. Two senior radiologists and one associate chief neurosurgeon jointly reviewed the original axial CTA images and performed image post-processing, documenting midline shift (mm) and the presence of intracranial hemorrhage, diffuse axonal injury, skull fracture, and brain edema.
Laboratory findings
Three months after the injury, a 5 ml blood sample was collected from the antecubital vein early in the morning while the patient was fasting. The levels of hemoglobin (g/dL), hematocrit (%), white blood cell count (×109/L), and platelet count (×109/L) were measured. The analyses were conducted using the STA Compact, China, a fully automated coagulation analyzer (model HC00608166). Additionally, arterial blood was sampled under fasting conditions and analyzed using the Roche automated biochemical analyzer with accompanying reagents, along with an automated blood gas analyzer to assess the patient’s biochemical indicators and perform arterial blood gas analysis, examining lactate, pH, and PaO2.
Treatment methods
Upon admission, patients received individualized treatment tailored to their specific conditions, and the interventions were duly documented. Over the three months post-admission, any complications experienced by the patients were meticulously recorded.
Statistical analysis
Data analysis was conducted using SPSS 29.0 statistical software (SPSS Inc., Chicago, IL, USA). Categorical data were presented as [n (%)] and analyzed using the chi-square test. Normality of continuous variables was assessed using the Shapiro-Wilk method. For normally distributed continuous variables, data were presented in the form of (mean ± SD) and analyzed using the t-test with corrected variance. A two-tailed P < 0.05 was considered statistically significant. The relationship between continuous variables, such as platelet count and PaO2 (mmHg), and prognostic outcomes was assessed using Pearson correlation analysis, while the relationship between categorical variables, such as the presence of subarachnoid hemorrhage and prognostic outcomes, was evaluated using Spearman correlation analysis. The diagnostic accuracy of PbtO2 differences in brain injury was assessed using the area under the receiver operating characteristic (ROC) curve (AUC). Variables showing significant differences in both difference and correlation analyses were included as covariates in logistic regression analysis.
Results
Demographic and clinical characteristics
The demographic and clinical characteristics of the study population are summarized in Table 1. There were 321 patients in the good prognosis group and 91 patients in the poor prognosis group. Age, gender, body mass index (BMI), smoking history, drinking history, hypertension, diabetes, and hyperlipidemia showed no significant differences between the two prognosis groups (P > 0.05 for all). However, the ISS and Admission GCS demonstrated statistically significant differences. Specifically, the poor prognosis group exhibited a significantly higher ISS (30.45 ± 7.32 vs. 26.89 ± 5.67, t = 4.938, P < 0.001) and lower GCS on admission (8.80 ± 3.20 vs. 11.02 ± 2.80, t = 6.462, P < 0.001) compared to the good prognosis group. These results suggest that ISS and GCS may play a pivotal role in predicting the neurological prognosis after TBI.
Table 1.
Comparison of demographic and clinical characteristics between the two groups
| Parameter | Good prognosis (n = 321) | Poor prognosis (n = 91) | t/χ2 | p |
|---|---|---|---|---|
| Age (years) | 39.14 ± 10.23 | 40.67 ± 12.85 | 1.186 | 0.236 |
| Gender (M/F) | 195 (60.75%)/126 (39.25%) | 57 (62.64%)/34 (37.36%) | 0.107 | 0.744 |
| BMI (kg/m2) | 23.98 ± 2.14 | 24.26 ± 1.92 | 1.126 | 0.261 |
| Smoking history (%) | 44 (13.71%) | 10 (10.99%) | 0.460 | 0.498 |
| Drinking history (%) | 26 (8.10%) | 10 (10.99%) | 0.742 | 0.389 |
| Hypertension (%) | 57 (17.76%) | 15 (16.48% | 0.080 | 0.778 |
| Diabetes (%) | 70 (21.81%) | 18 (19.78%) | 0.173 | 0.677 |
| Hyperlipidemia (%) | 48 (14.95%) | 12 (13.19%) | 0.178 | 0.673 |
| ISS | 26.89 ± 5.67 | 30.45 ± 7.32 | 4.938 | < 0.001 |
| Admission GCS | 11.02 ± 2.80 | 8.80 ± 3.20 | 6.462 | < 0.001 |
BMI: Body mass index; ISS: Injury severity score; GCS: Glasgow coma scale.
PbtO2 monitoring
A comparison of PbtO2 levels between the two groups revealed that initial PbtO2 levels were significantly higher in the good prognosis group than in the poor prognosis group (26.85 ± 4.82 mmHg vs. 20.34 ± 5.67 mmHg, t = 10.922, P < 0.001) (Table 2). Additionally, the proportion of patients with PbtO2 below 20 mmHg and 25 mmHg were significantly different between the two prognosis groups (PbtO2 < 20 mmHg: 9.03% vs. 23.08%, t = 13.112, P < 0.001; PbtO2 < 25 mmHg: 17.76% vs. 31.87%, t = 8.548, P = 0.003). However, there were no significant differences in the percentages of patients with PbtO2 measurements below 30 mmHg and 35 mmHg between the good and poor prognosis groups (P > 0.05). These findings underscore the potential influence of PbtO2 levels, particularly when below certain thresholds, on the neurological prognosis following TBI, warranting further investigation into the role of PbtO2 monitoring in this context.
Table 2.
Comparison of PbtO2 between the two groups
| Parameter | Good prognosis (n = 321) | Poor prognosis (n = 91) | t/χ2 | P |
|---|---|---|---|---|
| Initial PbtO2 (mmHg) | 26.85 ± 4.82 | 20.34 ± 5.67 | 10.922 | < 0.001 |
| PbtO2 < 20 mmHg (%) | 29 (9.03%) | 21 (23.08%) | 13.112 | < 0.001 |
| PbtO2 < 25 mmHg (%) | 57 (17.76%) | 29 (31.87%) | 8.548 | 0.003 |
| PbtO2 < 30 mmHg (%) | 165 (51.40%) | 55 (60.44%) | 2.327 | 0.127 |
| PbtO2 < 35 mmHg (%) | 195 (60.75%) | 58 (63.74%) | 0.267 | 0.605 |
PbtO2: Brain tissue oxygen pressure.
Radiological findings
The presence of diffuse axonal injury was significantly higher in the poor prognosis group compared to the good prognosis group (2.41 ± 1.20 vs. 1.72 ± 0.50, t = 8.126, P < 0.001) (Table 3). Furthermore, subarachnoid hemorrhage was more prevalent in the poor prognosis group compared to the good prognosis group (58.24% vs. 42.99%, t = 6.632, P = 0.010), while there were no significant differences in the incidence of intraventricular hemorrhage, base-of-skull fracture, and epidural hematoma between the two prognosis groups (all P > 0.05). These results indicate the potential impact of certain radiological findings, particularly diffuse axonal injury and subarachnoid hemorrhage, on neurological prognosis following TBI.
Table 3.
Comparison of radiological findings between the two groups
| Parameter | Good prognosis (n = 321) | Poor prognosis (n = 91) | t/χ2 | P |
|---|---|---|---|---|
| Diffuse axonal injury | 1.72 ± 0.50 | 2.41 ± 1.20 | 8.126 | < 0.001 |
| Subarachnoid hemorrhage | 138 (42.99%) | 53 (58.24%) | 6.632 | 0.010 |
| Intraventricular hemorrhage | 44 (13.71%) | 16 (17.58%) | 0.856 | 0.355 |
| Base-of-skull fracture | 80 (24.92%) | 27 (29.67%) | 0.831 | 0.362 |
| Epidural hematoma | 110 (34.27%) | 35 (38.46%) | 0.547 | 0.460 |
Laboratory findings
The platelet count was significantly lower in the poor prognosis group compared to the good prognosis group (198.4 ± 45.18×109/L vs. 230.7 ± 35.6×109/L, t = 7.174, P < 0.001), as well as a lower PaO2 (89.69 ± 7.29 mmHg vs. 93.62 ± 6.54 mmHg, t = 4.930, P < 0.001) (Figure 1). However, there were no significant differences in the levels of hemoglobin, hematocrit, white blood cell count, lactate, and arterial pH between the two prognosis groups (all P > 0.05). These findings suggest that certain laboratory parameters, particularly platelet count and PaO2, may be associated with neurological prognosis following TBI.
Figure 1.
Comparison of laboratory findings between the two groups of patients. Ns: no significant difference; ***: P < 0.001. A. Hemoglobin; B. Hematocrit; C. White blood cell count; D. Platelet count; E. Lactate; F. Arterial PH; G. PaO2. Notes: PH: Potential of hydrogen; PaO2: Arterial oxygen partial pressure.
Treatment interventions
There were no significant differences in the utilization of decompressive craniectomy, intracranial pressure management, hypertonic saline therapy, seizure prophylaxis, or hypothermia treatment between the two prognosis groups (all P > 0.05) (Table 4). These results suggest that the treatment interventions evaluated in this study may not be significantly associated with the neurological prognosis following TBI.
Table 4.
Comparison of treatment Interventions between the two groups
| Parameter | Good prognosis (n = 321) | Poor prognosis (n = 91) | χ2 | P |
|---|---|---|---|---|
| Decompressive craniectomy | 45 (14.02%) | 17 (18.68%) | 1.206 | 0.272 |
| Intracranial pressure management | 255 (79.44%) | 71 (78.02%) | 0.086 | 0.769 |
| Hypertonic saline therapy | 61 (19.00%) | 18 (19.78%) | 0.028 | 0.868 |
| Seizure prophylaxis | 181 (56.39%) | 49 (53.85%) | 0.186 | 0.667 |
| Hypothermia treatment | 23 (7.17%) | 11 (12.09%) | 2.269 | 0.132 |
Complications
There were no significant differences in the occurrences of seizure, permanent need for ventilator support, deep vein thrombosis, or inpatient mortality between the two prognosis groups (all P > 0.05) (Table 5). These findings suggest that the investigated complications may not have a significant association with the neurological prognosis following TBI.
Table 5.
Comparison of complications between the two groups
| Parameter | Good prognosis (n = 321) | Poor prognosis (n = 91) | χ2 | P |
|---|---|---|---|---|
| Seizure | 21 (6.54%) | 8 (8.79%) | 0.548 | 0.459 |
| Permanently needing ventilator | 43 (13.40%) | 15 (16.48%) | 0.559 | 0.455 |
| Deep vein thrombosis | 20 (6.23%) | 5 (5.49%) | 0.067 | 0.795 |
| Inpatient mortality | 5 (1.56%) | 3 (3.30%) | 1.126 | 0.289 |
Correlation analysis
The ISS demonstrated a positive correlation (r = 0.225, P < 0.001), while admission GCS showed a negative correlation (r = -0.275, P < 0.001) with poor neurological outcomes (Figure 2). Additionally, initial PbtO2, PaO2 and platelet count exhibited negative correlations (r = -0.437, P < 0.001; r = -0.236, P < 0.001; and r = -0.300, P < 0.001, respectively) with poor neurological outcomes. Subarachnoid hemorrhage, PbtO2 below 25 mmHg, PbtO2 below 20 mmHg and diffuse axonal injury also demonstrated positive correlations (r = 0.127, P < 0.001; r = 0.144, P < 0.001; r = 0.178, P < 0.001 and r = 0.277, P < 0.001, respectively) with poor neurological outcomes. The correlation analysis provides valuable insight into the relationships between these influencing factors and poor neurological outcomes.
Figure 2.
Correlation analysis between various influencing factors and poor neurological outcome after TBI. A. Correlation coefficient; B. P value. Notes: TBI: Traumatic brain injury; PbtO2: Brain tissue oxygen pressure; PaO2: Arterial oxygen partial pressure; GCS: Glasgow coma scale.
Logistic regression analysis
Logistic regression analysis shows that a higher ISS (ISS ≥ 33.53, OR: 1.101, P < 0.001) is significantly associated with an increased risk of poor neurological outcomes (Table 6). Conversely, a higher GCS score at admission (GCS ≥ 8.7, OR: 0.762, P < 0.001) is significantly associated with a decreased risk of poor outcomes. Similarly, a higher initial PbtO2 (PbtO2 ≥ 25.04, OR: 0.772, P < 0.001) is associated with a decreased risk. The presence of diffuse axonal injury (diffuse axonal injury ≥ 2.42, OR: 3.310, P < 0.001) is significantly associated with an increased risk of poor outcomes. A higher platelet count (Platelet Count ≥ 194.31×10^9/L, OR: 0.979, P < 0.001) is associated with a decreased risk. A higher PaO2 (PaO2 ≥ 89.38, OR: 0.916, P < 0.001) is also associated with a decreased risk. Lower PbtO2 (PbtO2 < 20 mmHg, OR: 3.021, P < 0.001 and PbtO2 < 25 mmHg, OR: 2.166, P = 0.004) is significantly associated with increased risks. Finally, the presence of subarachnoid hemorrhage (OR: 1.850, P = 0.011) is significantly associated with an increased risk. These findings highlight the importance of these clinical parameters in predicting and managing the risk of poor outcomes in TBI patients.
Table 6.
Logistic regression analysis of various factors for poor neurological outcome post-TBI
| Parameter | Coef | Odds ratio | B | beta | P |
|---|---|---|---|---|---|
| ISS ≥ 33.53 | 0.097 | 1.101 | 4.668 | 0.097 | < 0.001 |
| Admission GCS ≥ 8.7 | 0.271 | 0.762 | 5.838 | -0.271 | < 0.001 |
| Initial PbtO2 ≥ 25.04 | 0.259 | 0.772 | 8.27 | -0.259 | < 0.001 |
| Diffuse axonal injury ≥ 2.42 | 1.197 | 3.310 | 6.583 | 1.197 | < 0.001 |
| Platelet count ≥ 194.31×109/L | 0.021 | 0.979 | 6.338 | -0.021 | < 0.001 |
| PaO2 ≥ 89.38 | 0.088 | 0.916 | 4.652 | -0.088 | < 0.001 |
| PbtO2 < 20 mmHg (%) | 1.105 | 3.021 | 3.499 | 1.105 | < 0.001 |
| PbtO2 < 25 mmHg (%) | 0.773 | 2.166 | 2.882 | 0.773 | 0.004 |
| Subarachnoid hemorrhage | 0.615 | 1.850 | 2.556 | 0.615 | 0.011 |
TBI: Traumatic brain injury; ISS: Injury severity score; GCS: Glasgow coma scale; PbtO2: Brain tissue oxygen pressure.
ROC
ROC analysis assessed the predictive value of various influencing factors for poor neurological outcome after TBI, revealing several parameters with meaningful predictive potential (Table 7). Specifically, initial PbtO2 demonstrated the highest AUC at 0.804 (Figure 3), indicating strong predictive value for poor neurological outcomes. Additionally, admission GCS and platelet count also exhibited notable AUC values of 0.692 and 0.709, respectively. While ISS, PaO2, and diffuse axonal injury showcased moderate AUC values of 0.657, 0.665, and 0.693, respectively. Although PbtO2 below 20 mmHg, PbtO2 below 25 mmHg, and the presence of subarachnoid hemorrhage demonstrated lower AUC values, they still provide some predictive value for poor neurological outcomes. The above results show that the initial PbtO2 in predicting adverse neurological prognosis after TBI has potential utility.
Table 7.
The predictive value of various influencing factors for poor neurological outcome after TBI
| Parameter | Sensitivities | Specificities | AUC | Youden index |
|---|---|---|---|---|
| ISS | 0.418 | 0.872 | 0.657 | 0.290 |
| Admission GCS | 0.505 | 0.801 | 0.692 | 0.306 |
| Initial PbtO2 (mmHg) | 0.813 | 0.648 | 0.804 | 0.461 |
| Diffuse axonal injury | 0.516 | 0.916 | 0.693 | 0.432 |
| Platelet count (×109/L) | 0.495 | 0.844 | 0.709 | 0.339 |
| PaO2 (mmHg) | 0.571 | 0.735 | 0.665 | 0.306 |
| PbtO2 < 20 mmHg (%) | 0.231 | 0.910 | 0.570 | 0.141 |
| PbtO2 < 25 mmHg (%) | 0.319 | 0.822 | 0.571 | 0.141 |
| Subarachnoid hemorrhage | 0.582 | 0.570 | 0.576 | 0.152 |
TBI: Traumatic brain injury; ISS: Injury severity score; GCS: Glasgow coma scale; PbtO2: Brain tissue oxygen pressure.
Figure 3.
The predictive value of initial PbtO2 for poor neurological outcome post-TBI. Notes: TBI: Traumatic brain injury; PbtO2: Brain tissue oxygen pressure.
Discussion
Traumatic brain injury (TBI) remains a major public health concern due to its profound impact on individuals and society. Understanding the factors that influence neurological prognosis following TBI is crucial for improving patient outcomes and guiding clinical decision-making [23-25].
The demographic and clinical characteristics of the study population revealed several noteworthy findings. Although age, gender, BMI, smoking history, alcohol consumption, hypertension, and diabetes did not significantly differ between patients with good and poor prognosis, ISS and admission GCS demonstrated clear associations with neurological outcomes. The ISS represents the severity of injuries across different body regions, whereas the GCS assesses the level of consciousness. As evidenced by our findings, a higher ISS and lower GCS on admission were significantly associated with poor neurological prognoses, aligning with existing literature on the predictive value of these scoring systems. The association of high ISS and low GCS with poor prognosis underscores the critical role of injury severity and initial consciousness level in determining neurological outcomes after TBI [26,27]. These results emphasize the importance of early and accurate assessment of injury severity and consciousness level for prognostication and targeted management strategies post-TBI.
The evaluation of PbtO2 levels revealed intriguing findings, with initial PbtO2 levels significantly higher in patients with good prognosis than in those with poor prognosis. Moreover, the percentage of PbtO2 below certain thresholds (20 mmHg and 25 mmHg) differed significantly between the two prognosis groups. These findings suggest that PbtO2 monitoring may offer valuable insights into managing cerebral oxygenation status post-TBI. The association of lower PbtO2 levels with poorer prognosis aligns with the established role of cerebral hypoxia in TBI outcomes [28,29]. It was well-established that maintaining adequate cerebral oxygenation is critical for minimizing secondary brain injury and enhancing recovery following TBI. The observed association between PbtO2 levels and neurological prognosis underscores the potential utility of PbtO2 monitoring as a prognostic tool in TBI management. Notably, the correlation and logistic regression analyses highlight the independent predictive value of PbtO2 levels, particularly when they fall below certain thresholds, in determining neurological outcomes. The robust predictive potential of initial PbtO2 levels, as indicated by AUC, underscores the clinical relevance of PbtO2 monitoring in assessing and predicting neurological prognoses following TBI.
The radiological findings in our study revealed significant associations between certain imaging parameters and neurological prognoses. Specifically, the presence of diffuse axonal injury and subarachnoid hemorrhage demonstrated clear associations with poor neurological outcomes. Diffuse axonal injury, characterized by widespread damage to axonal fibers, was associated with cognitive and functional impairments [30-32], while subarachnoid hemorrhage can lead to increased intracranial pressure and cerebral complications [33-35]. The correlation of these radiological findings with poor neurological prognoses underscores the critical role of comprehensive neuroimaging assessments in understanding and predicting TBI outcomes. The identification of these specific radiological markers as independent risk factors for poor neurological prognosis emphasizes the need for meticulous radiological evaluation and interpretation in TBI management.
The analysis of laboratory parameters revealed notable associations between platelet count, PaO2, and neurological prognoses. Specifically, lower platelet counts and PaO2 levels were significantly associated with poor neurological outcomes. These findings align with the recognized impact of coagulation abnormalities and hypoxemia on TBI outcomes. Impaired platelet function and low PaO2 levels can exacerbate secondary brain injury and negatively influence TBI recovery processes [36-38]. The association of these laboratory parameters with neurological outcomes highlights the potential utility of comprehensive laboratory assessments in prognostication and risk stratification following TBI. The predictive potential of platelet count and PaO2 levels, as indicated by their AUC values in the ROC analysis, further emphasizes their clinical relevance in predicting neurological prognoses post-TBI.
The findings of this study underscore the multifactorial nature of prognostication in traumatic brain injury, aligning with existing literature [39] that emphasizes a comprehensive approach to evaluating TBI outcomes. Previous studies [40,41] have consistently highlighted the predictive value of the GCS and ISS in assessing the initial severity of brain injuries. Our study corroborates these findings, demonstrating significant correlations between lower GCS and higher ISS scores with poor neurological outcomes [42]. These tools remain vital in the clinical setting, serving as reliable indicators of initial injury severity and facilitating early intervention planning.
While the present study offers valuable insights into the factors influencing neurological prognosis after TBI, several limitations should be acknowledged. The retrospective nature and reliance on de-identified patient data may have introduced inherent biases and limitations in data collection and analysis. Prospective studies with larger sample sizes and extended follow-up periods are needed to validate and expand upon the current findings. Additionally, the single-center design may limit the generalizability of the results. Multi-center studies involving diverse patient populations and settings are warranted to corroborate the identified influencing factors and their associations with TBI prognosis.
Conclusion
In conclusion, this study underscores the prognostic significance of injury severity, brain tissue oxygenation, radiological findings, and laboratory parameters in determining neurological outcomes following TBI. These results deepen the understanding of the multifaceted determinants of TBI prognosis and suggest a potential role for PbtO2 monitoring as a valuable tool in prognostic assessment. Further exploration and validation in prospective, multi-center studies are essential to refine prognostic models and enhance clinical decision-making in TBI management. Ultimately, a comprehensive evaluation of influencing factors identified in this study may contribute to the development of tailored, evidence-based strategies to optimize neurological prognosis and outcomes in TBI patients.
Disclosure of conflict of interest
None.
References
- 1.Bailes JE, Borlongan CV. Traumatic brain injury. CNS Neurosci Ther. 2020;26:593–594. doi: 10.1111/cns.13397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Khellaf A, Khan DZ, Helmy A. Recent advances in traumatic brain injury. J Neurol. 2019;266:2878–2889. doi: 10.1007/s00415-019-09541-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Mostert CQB, Singh RD, Gerritsen M, Kompanje EJO, Ribbers GM, Peul WC, van Dijck JTJM. Long-term outcome after severe traumatic brain injury: a systematic literature review. Acta Neurochir (Wien) 2022;164:599–613. doi: 10.1007/s00701-021-05086-6. [DOI] [PubMed] [Google Scholar]
- 4.Valadka AB, Maas AIR, Servadei F. Introduction. Traumatic brain injury. Neurosurg Focus. 2019;47:E1. doi: 10.3171/2019.8.FOCUS19688. [DOI] [PubMed] [Google Scholar]
- 5.Ritter M. Evidence-based pearls: traumatic brain injury. Crit Care Nurs Clin North Am. 2023;35:171–178. doi: 10.1016/j.cnc.2023.02.009. [DOI] [PubMed] [Google Scholar]
- 6.Robinson CP. Moderate and severe traumatic brain injury. Continuum (Minneap Minn) 2021;27:1278–1300. doi: 10.1212/CON.0000000000001036. [DOI] [PubMed] [Google Scholar]
- 7.Pinggera D, Geiger P, Thomé C. Traumatic brain injury. Nervenarzt. 2023;94:960–972. doi: 10.1007/s00115-023-01546-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Leikin JB. Traumatic brain injury. Dis Mon. 2019;65:100857. doi: 10.1016/j.disamonth.2019.02.010. [DOI] [PubMed] [Google Scholar]
- 9.de Cássia Almeida Vieira R, Silveira JCP, Paiva WS, de Oliveira DV, de Souza CPE, Santana-Santos E, de Sousa RMC. Prognostic models in severe traumatic brain injury: a systematic review and meta-analysis. Neurocrit Care. 2022;37:790–805. doi: 10.1007/s12028-022-01547-7. [DOI] [PubMed] [Google Scholar]
- 10.Smith LGF, Milliron E, Ho ML, Hu HH, Rusin J, Leonard J, Sribnick EA. Advanced neuroimaging in traumatic brain injury: an overview. Neurosurg Focus. 2019;47:E17. doi: 10.3171/2019.9.FOCUS19652. [DOI] [PubMed] [Google Scholar]
- 11.Sharma B, Changoor A, Monteiro L, Colella B, Green R. Prognostic-factors for neurodegeneration in chronic moderate-to-severe traumatic brain injury: a systematic review protocol. Syst Rev. 2020;9:23. doi: 10.1186/s13643-020-1281-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Parker TD, Rees R, Rajagopal S, Griffin C, Goodliffe L, Dilley M, Jenkins PO. Post-traumatic amnesia. Pract Neurol. 2022;22:129–137. doi: 10.1136/practneurol-2021-003056. [DOI] [PubMed] [Google Scholar]
- 13.Baxendale S, Heaney D, Rugg-Gunn F, Friedland D. Neuropsychological outcomes following traumatic brain injury. Pract Neurol. 2019;19:476–482. doi: 10.1136/practneurol-2018-002113. [DOI] [PubMed] [Google Scholar]
- 14.Hawryluk GWJ, Rubiano AM, Totten AM, O’Reilly C, Ullman JS, Bratton SL, Chesnut R, Harris OA, Kissoon N, Shutter L, Tasker RC, Vavilala MS, Wilberger J, Wright DW, Lumba-Brown A, Ghajar J. Guidelines for the management of severe traumatic brain injury: 2020 update of the decompressive craniectomy recommendations. Neurosurgery. 2020;87:427–434. doi: 10.1093/neuros/nyaa278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Retel Helmrich IRA, Lingsma HF, Turgeon AF, Yamal JM, Steyerberg EW. Prognostic research in traumatic brain injury: markers, modeling, and methodological principles. J Neurotrauma. 2021;38:2502–2513. doi: 10.1089/neu.2019.6708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sabet N, Soltani Z, Khaksari M. Multipotential and systemic effects of traumatic brain injury. J Neuroimmunol. 2021;357:577619. doi: 10.1016/j.jneuroim.2021.577619. [DOI] [PubMed] [Google Scholar]
- 17.Scarboro M, McQuillan KA. Traumatic brain injury update. AACN Adv Crit Care. 2021;32:29–50. doi: 10.4037/aacnacc2021331. [DOI] [PubMed] [Google Scholar]
- 18.Thapa K, Khan H, Singh TG, Kaur A. Traumatic brain injury: mechanistic insight on pathophysiology and potential therapeutic targets. J Mol Neurosci. 2021;71:1725–1742. doi: 10.1007/s12031-021-01841-7. [DOI] [PubMed] [Google Scholar]
- 19.Gao G, Wu X, Feng J, Hui J, Mao Q, Lecky F, Lingsma H, Maas AIR, Jiang J China CENTER-TBI Registry Participants. Clinical characteristics and outcomes in patients with traumatic brain injury in China: a prospective, multicentre, longitudinal, observational study. Lancet Neurol. 2020;19:670–677. doi: 10.1016/S1474-4422(20)30182-4. [DOI] [PubMed] [Google Scholar]
- 20.Maragkos GA, Cho LD, Legome E, Wedderburn R, Margetis K. Prognostic factors for stage 3 acute kidney injury in isolated serious traumatic brain injury. World Neurosurg. 2022;161:e710–e722. doi: 10.1016/j.wneu.2022.02.106. [DOI] [PubMed] [Google Scholar]
- 21.Rodriguez A, Edelen MO, Herman PM, Hays RD. Unpacking the impact of chronic pain as measured by the impact stratification score. BMC Musculoskelet Disord. 2022;23:884. doi: 10.1186/s12891-022-05834-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Caruana M, Hackenbruch SN, Grech V, Farrugia R. Inconsistency in the application of glasgow coma scale in pediatric patients. Med Princ Pract. 2024;33:41–46. doi: 10.1159/000534797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kanthimathinathan HK, Mehta H, Scholefield BR, Morris KP. Traumatic brain injury practice guidelines: variability in U.K. PICUs. Pediatr Crit Care Med. 2021;22:e270–e274. doi: 10.1097/PCC.0000000000002574. [DOI] [PubMed] [Google Scholar]
- 24.Ladak AA, Enam SA, Ibrahim MT. A review of the molecular mechanisms of traumatic brain injury. World Neurosurg. 2019;131:126–132. doi: 10.1016/j.wneu.2019.07.039. [DOI] [PubMed] [Google Scholar]
- 25.Chinnery PF. Traumatic brain injury advances since 2017: what has changed? Lancet Neurol. 2022;21:953–954. doi: 10.1016/S1474-4422(22)00337-4. [DOI] [PubMed] [Google Scholar]
- 26.Van Deynse H, Cools W, Depreitere B, Hubloue I, Kazadi CI, Kimpe E, Moens M, Pien K, Van Belleghem G, Putman K. Quantifying injury severity for traumatic brain injury with routinely collected health data. Injury. 2022;53:11–20. doi: 10.1016/j.injury.2021.10.013. [DOI] [PubMed] [Google Scholar]
- 27.Li A, Atem FD, Venkatachalam AM, Barnes A, Stutzman SE, Olson DM. Admission glasgow coma scale score as a predictor of outcome in patients without traumatic brain injury. Am J Crit Care. 2021;30:350–355. doi: 10.4037/ajcc2021163. [DOI] [PubMed] [Google Scholar]
- 28.Bernard F, Barsan W, Diaz-Arrastia R, Merck LH, Yeatts S, Shutter LA. Brain oxygen optimization in severe traumatic brain injury (BOOST-3): a multicentre, randomised, blinded-endpoint, comparative effectiveness study of brain tissue oxygen and intracranial pressure monitoring versus intracranial pressure alone. BMJ Open. 2022;12:e060188. doi: 10.1136/bmjopen-2021-060188. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Chesnut R, Aguilera S, Buki A, Bulger E, Citerio G, Cooper DJ, Arrastia RD, Diringer M, Figaji A, Gao G, Geocadin R, Ghajar J, Harris O, Hoffer A, Hutchinson P, Joseph M, Kitagawa R, Manley G, Mayer S, Menon DK, Meyfroidt G, Michael DB, Oddo M, Okonkwo D, Patel M, Robertson C, Rosenfeld JV, Rubiano AM, Sahuquillo J, Servadei F, Shutter L, Stein D, Stocchetti N, Taccone FS, Timmons S, Tsai E, Ullman JS, Vespa P, Videtta W, Wright DW, Zammit C, Hawryluk GWJ. A management algorithm for adult patients with both brain oxygen and intracranial pressure monitoring: the seattle international severe traumatic brain injury consensus conference (SIBICC) Intensive Care Med. 2020;46:919–929. doi: 10.1007/s00134-019-05900-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Jang SH. Diagnostic problems in diffuse axonal injury. Diagnostics (Basel) 2020;10:117. doi: 10.3390/diagnostics10020117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Krieg JL, Leonard AV, Turner RJ, Corrigan F. Identifying the phenotypes of diffuse axonal injury following traumatic brain injury. Brain Sci. 2023;13:1607. doi: 10.3390/brainsci13111607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Mesfin FB, Gupta N, Hays Shapshak A, Taylor RS. Diffuse Axonal Injury. In: StatPearls. Treasure Island (FL) ineligible companies. Disclosure: nishant gupta declares no relevant financial relationships with ineligible companies. Disclosure: angela hays shapshak declares no relevant financial relationships with ineligible companies. Disclosure: roger taylor declares no relevant financial relationships with ineligible companies.: StatPearls Publishing. Copyright © 2024, StatPearls Publishing LLC.; 2024. [Google Scholar]
- 33.Griswold DP, Fernandez L, Rubiano AM. Traumatic subarachnoid hemorrhage: a scoping review. J Neurotrauma. 2022;39:35–48. doi: 10.1089/neu.2021.0007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hoh BL, Ko NU, Amin-Hanjani S, Chou SH-Y, Cruz-Flores S, Dangayach NS, Derdeyn CP, Du R, Hänggi D, Hetts SW, Ifejika NL, Johnson R, Keigher KM, Leslie-Mazwi TM, Lucke-Wold B, Rabinstein AA, Robicsek SA, Stapleton CJ, Suarez JI, Tjoumakaris SI, Welch BG. 2023 guideline for the management of patients with aneurysmal subarachnoid hemorrhage: a guideline from the American heart association/American stroke association. Stroke. 2023;54:e314–e370. doi: 10.1161/STR.0000000000000436. [DOI] [PubMed] [Google Scholar]
- 35.Karceski S. Subarachnoid hemorrhage: the long and short of it. Neurology. 2020;95:e1915–e1917. doi: 10.1212/WNL.0000000000010642. [DOI] [PubMed] [Google Scholar]
- 36.Dong JF, Zhang F, Zhang J. Detecting traumatic brain injury-induced coagulopathy: what we are testing and what we are not. J Trauma Acute Care Surg. 2023;94(Suppl 1):S50–S55. doi: 10.1097/TA.0000000000003748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Liu L, Deng QJ. Role of platelet-derived extracellular vesicles in traumatic brain injury-induced coagulopathy and inflammation. Neural Riojas CM, Ekaney ML, Ross SW, Cunningham KW, Furay EJ, Brown CVR and Evans SL. Platelet dysfunction after traumatic brain injury: a review. J Neurotrauma. 2021;38:819–829. doi: 10.1089/neu.2020.7301. [DOI] [PubMed] [Google Scholar]
- 38.Riojas CM, Ekaney ML, Ross SW, Cunningham KW, Furay EJ, Brown CVR, Evans SL. Platelet dysfunction after traumatic brain injury: a review. J Neurotrauma. 2021;38:819–829. doi: 10.1089/neu.2020.7301. [DOI] [PubMed] [Google Scholar]
- 39.Maas AIR, Hemphill JC, Wilson L, Manley GT. Managing outcome expectations after traumatic brain injury. Injury. 2023;54:1233–1235. doi: 10.1016/j.injury.2023.03.027. [DOI] [PubMed] [Google Scholar]
- 40.Rafay M, Gulzar F, Iqbal N, Sharif S. Prognostic computed tomography scores in traumatic brain injury. Clin Neurol Neurosurg. 2020;195:105957. doi: 10.1016/j.clineuro.2020.105957. [DOI] [PubMed] [Google Scholar]
- 41.Tunthanathip T, Phuenpathom N, Jongjit A. Prognostic factors and clinical nomogram for in-hospital mortality in traumatic brain injury. Am J Emerg Med. 2024;77:194–202. doi: 10.1016/j.ajem.2023.12.037. [DOI] [PubMed] [Google Scholar]
- 42.Capizzi A, Woo J, Verduzco-Gutierrez M. Traumatic brain injury: an overview of epidemiology, pathophysiology, and medical management. Med Clin North Am. 2020;104:213–238. doi: 10.1016/j.mcna.2019.11.001. [DOI] [PubMed] [Google Scholar]