The Outcome Predictors of the Patients with Traumatic Brain Injury; A Cross-Sectional Study

Introduction

Traumatic Brain Injury (TBI) is a common cause of neurological disability and death. It is estimated that about 1.5 million people die worldwide following TBI. TBI also is one of the leading causes of mortality in the intensive care units (ICU) and emergency department (ED) of major trauma centers [1].

An accurate assessment of prognosis after TBI is very important to make decisions of using specific treatment’s method, preventing nosocomial infections, counseling patients and relatives, and identify the specific rehabilitation of the patient needs [2, 3]. Most patients with severe TBI are unconscious, intubated, anesthetized, and sedated, which makes the use of clinical evaluation of the injury severity less reliable like applying the Glasgow Coma Scale (GCS) [4]. Therefore, mostly required undergoing emergent computed tomography (CT) scan is as a part of their secondary survey in the ED for providing information on the patients’ outcome prediction [5].

In previous studies, some factors have been linked to increase mortality and worse outcome after TBI such as age, the coexistence of other injuries, history of previous head injury, alcohol and drug abuse, low socioeconomic and educational status [6, 7]. Available data revealed that the risk of TBI is high in the 15-24 years of age that decrease in the midlife, and then increases again after 70 years which mostly due to falls. Men sustain a TBI approximately 3 to 4 times as likely as women, but this ratio narrows in the elderly [8]. Half of fatal and non-fatal TBIs are due to motor-vehicle-collisions (MVCs), whereas the 2^nd most frequent cause of TBI is falls [9].

These findings are often derived from studies conducted in developed countries and less information are available from other societies. In developing and the third world countries, the societies’ characteristics are noticeably different from developed countries such as the job’s type and frequency, the health system, the quality of cars, traffic laws and many other factors. Therefore, there is a need to conduct studies on a large scale in these communities. In this study, our purpose was to evaluate the epidemiological aspects of TBI patients and also to probe the affecting factors on the patients’ outcomes with TBI who admitted to ED.

Materials and Methods

This was a cross-sectional study conducted in Tehran, Iran. Data gathering was performed via census manner. We retrospectively recruited TBI patients of all ages and gender with any level of consciousness who admitted to the ED of Sina hospital, Tehran, Iran during 2017-2019. Those with distorted evidence or incomplete data were excluded.

The required data were extracted by using a pre-prepared checklist for further analysis. We were recorded age (further categorized as <4, 5-14, 15-24, 25-44, 45-64 and ≥65 years), gender, mechanism of injury, Glasgow coma scale (GCS) on admission, ISS (Injury Severity Score), presence of extra-cranial injuries, CT scan findings, duration of hospitalization, and in hospital outcomes. Therefore, the required information was collected in advance by trained experts in the form of “National Trauma Registry of Iran”. Although, outcome assessment was performed for the survivors within a 6 months-period after discharge based on Glasgow outcome scale (GOS). Unfavorable outcome (death or severe disability) was defined in six months with the Glasgow outcome scale (GOS). The scale comprises five categories: death, vegetative state, severe disability, moderate disability, low disability and good recovery.

Patients’ data was collected and analyzed statistically with SPSS software (IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp; 2016). Logistic regression was used to study the effective factors as univariate and multivariate. Chi-square analysis, analysis of variance and comparison of means were used to investigate the effect of each of the independent factors and variables with the outcome according to the nature of the variable.

Results

During the one-year study period, 506 patients were admitted due to TBI. Baseline characteristics of the study patients are presented in Table 1. A preponderance of injuries occurred among men (81.2%) and the age between 25-44 years (35.6%). The minimum and maximum age of the patients was 1 and 94 years, respectively. The mean age of the patients was 36.77 (SD: 21.1) years which was 36.27 (SD: 20.4) years for men and 38.98 (SD: 24.0) years for women. Mean corpuscular volume (MVCs) were the most frequent leading cause of TBI (73.99%), mostly in the youngest, followed by falls injury (20.6%) which more pronounced in the elderly population. Subarachnoid hemorrhage followed by epidural hematoma were the most frequent abnormal findings in brain CT scan of the patients. The ISS mean of the patients was 6.99 ± 8.28 and the proportion of the patients did not differ significantly in all injury severity categories and was almost equal. Follow up and outcome assessment was feasible in 487 (96.2%) patients after 6 months. According to GOS, patients’ outcome post-injury revealed 59 (11.7%) of cases who were in emerged death at six-months while good recovery was seen in 397 (78.5%) patients, and we missed 19 (3.8) cases at this step.

The univariate regression analysis showed that below variables had significant relation with the outcome (Table 2): age categories of 45-64 (p=0.030) and ≥65 (p=0.003); GCS<14 (p<0.0001); sever ISS (p<0.0001); presence of ICH (p<0.0001), subarachnoid hemorrhage (p=0.029) and subdural hematoma (p<0.0001) in brain CT scan.

Table 3 shows the results of multivariate regression analysis on mortality rate distribution by assessed variables. Patients above 65-years-old had worse outcome in comparing with the other age groups [adjusted OR: 12.21 (4.48, 33.24)]. As expected, GCS on admission was highly associated with mortality [adjusted OR: 62.99 (23.28, 170.46) for GCS 3-8]. Presence of abnormal findings in patients with brain CT scan was also another associated factor with mortality [adjusted OR: 20.11 (2.03, 199.27)]; and finally hospital length of stay more than 5 days was also associated with mortality [adjusted OR: 0.28 (0.12, 0.62)]. Gender, ISS and mechanism of injury did not show any significant association in this regard. Meanwhile, the proportion of isolated TBI patients were higher than those who suffered from multiple trauma, but we did not find that presence of extra-cranial injuries could aggravated the prognosis of the patients.

Discussion

A preponderance of injuries occurred among middle aged men. MVCs were the most frequent leading cause of TBI, mostly in the youngest patients, followed by falls’ injury that were more pronounced in the elderly population. Subarachnoid hemorrhage was the most frequent findings in patients’ CT scan. We reached the 11% mortality among our study patients. Elderly patients had worse outcome compared to the others. As expected, GCS on admission was highly associated with mortality. In general, logistic regression analysis showed that older age, lower GCS, presence of abnormal CT scan findings and duration of hospitalization more than 5 days were associated with the outcome of the patients with TBI.

Previous studies revealed that demographics characteristics, mechanism of injury, GCS and abnormal brain CT scan findings may affect the outcome of TBI patients [1, 10-12]. Gender does not have statistical significance in predicting outcome, although men are more likely to sustain a TBI than women [1, 5, 10]. This has been attributed to more men being drivers and involved in MVCs [13]. These are in line with the findings of the present study, except for mechanism of injury that did not show any association with the outcome.

In the present study, the overall mortality rate is lower than other studies, usually varying between 32-49% [5, 14, 15]. One of the reason is that the present study was conducted in a single center. But more importantly, the recorded information was used in a registry department. It is necessary to explain that this registry does not include patients who died before being transferred to inpatient departments in the early hours of ED admission, or cases of pre-hospital death.

The GCS score was described in 1974 by Teasdale and Jennett. They were assessed the degree of unconsciousness in patients with traumatic brain injury [16, 17]. Evidence shows that GCS is a strong predictor of outcome in TBI [1, 10]. However, it may be affected by sedation, paralysis or intoxication with alcohol and affected by presence of facial swelling [5, 18]. As expected, the results of the present study also showed that the mortality rate of the patients with lower on admission GCS, was higher than the other TBI patients.

Brain CT scan plays a crucial role in early assessment of patients with TBI. In poor resource settings, CT scan findings may be used as an important tool to predict the TBI patients’ outcome where intracranial pressure monitoring is not readily available [11]. In the present study, we report the highest mortality rate in patients with findings of subdural hematoma, while the most frequent image finding was intracranial hemorrhage.

If we want to speak about the role and impact of hospitalization length on the outcome of TBI patients, we must consider various aspects. It is expected that high-risk TBI patients, especially those with a lower level of consciousness at the time of ED admission or those with abnormal brain CT scan findings, will be admitted to the intensive care unit (ICU), and if they pass the critical phase, they will be transferred to the ward units, while they may also undergo one or more surgical interventions [19, 20]. On the other hand, the length of stay prolonged in the hospital, especially in the ICUs is associated with some problems such as the possibility of hospital infections affect, ventilator-associated pneumonia in intubated patients, occurring of deep venous thromboembolism and many other issues which will prolong the hospitalization duration of high risk in TBI patients [21-23]. Due to the existing restrictions in some parts of the world like Iran, it may not be possible to quickly transfer high-risk TBI patients to the ICU which is another important points, and the patients will be cared for a period of time in the ED which may sometimes last more than a few days. This is the maximum care that can be performed inevitably, but we believe that the quality of care in ED cannot be compared with the ICU. Of course, this is a hypothesis and perhaps a study should be conducted to show that delay in transferring patients from ED to ICU or other inpatient departments can affect the outcome of high risk TBI patients.

There may be require to develop a specific evidence-based guideline or at least a national protocol to manage high risk TBI patients in ED by considering the current study results and also the same studies conducted in the same era in various parts of the world including in developing countries.

In conclusion, we suggest to perform a systematic literature review for finding the risk factors and thereafter, distinguish the effective interventions that could alter the high risk TBI patient’s outcome. Albeit, increasing the level of person’s awareness, establishing and implementing rules will probably be very beneficial for using of safety devices and to prevent injuries following MVCs.

Limitations

This was a single center study in which we were used the recorded information of “Iran National Trauma Registry”. It is necessary to explain that this registry does not include patients who died before being transferred to inpatient departments in the early hours of ED admission, or cases of pre-hospital death. Therefore, and due to the retrospective nature of the study, an important part of the information of TBI patients was not available. In the future studies, fixing this shortcoming may bring different and at the same time more valuable results.

Declarations

Ethics approval and consent to participate: This study was approved by ethical committee of Tehran University of Medical Sciences (IR.TUMS.MEDICINE.REC.1397.289). The extracted data were recorded, analyzed and reported anonymously.

Consent for publication: None declared.

Conflict of interests: The authors declared that there is no conflict of interest.

Funding: This study was funded with a grant from Research Center for Trauma in Police Operation. The project was commissioned by the organization and the cost was paid to the contractor.

Authors’ contributions: The conception and design of the work by SR, AS and AB; Data acquisition by SR, AJ and KK; Analysis and interpretation of data by SR and AS; Drafting the work by SR and AJ; Revising it critically for important intellectual content by AS, KK and AB; All the authors approved the final version to be published; AND agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work.

Acknowledgements: We would like to express our commitment to the Sina Trauma and Surgery Research Center affiliated to Tehran University of Medical Sciences.