According to the World Health Organization, traumatic brain injury (TBI) is one of the most common causes of death and disability in the world, with more than 10 million hospitalizations annually (1, 2). Epidemiological data from the United States indicate 1.4 million TBI cases per year (3). For European populations, the TBI incidence is 235 per 100,000 per year, with a mortality rate of 0.015% (4). Traumatic brain injury is particularly common among people below 25 and above 75 years of age (5, 6), and in most of the reviewed studies, it is more prevalent among men than women (7, 8). The most common causes of traumatic brain injury in European populations include falls, mostly in children and elderly people, and road accidents, which are the leading cause of TBI among young adults. Additional causes include battery and sports- and recreation-related injuries (4).

TBI is a significant medical problem because it is characterized by high short-term mortality and a high risk of severe neurological complications in long-term follow-ups, and there is the risk of chronic cognitive and behavioral disorders, consciousness disorders, chronic traumatic encephalopathy, physical disability, psychiatric and neurodegenerative diseases, and organ dysfunction secondary to neurological deficits (9–12).

The assessment methods for brain injury are complex and presented as classification scales, such as the Abbreviated Injury Scale (AIS), Injury Severity Scale (ISS), Revised Trauma Score (RTS), or local injury assessments (the Mayo Classification System for TBI), and they are based on general neurological examinations and imaging scans (13, 14).

The Mayo Classification System for TBI is a typical TBI classification system. Possible TBI is connected with neurocognitive symptoms, such as blurred vision, confusion, headache, nausea, and a loss of consciousness for <30 min. According to the Mayo Classification System for TBI, a TBI is classified as mild (probable) if one or more of the following criteria apply: the loss of momentary consciousness to <30 min, the post-traumatic anterograde amnesia of momentary consciousness to <24 h, depression, and basilar or linear skull fracture (with the dura intact). The most serious TBI cases are moderate and severe TBI, according to the Mayo Classification System for TBI, and they are recognized when one or more of the following criteria apply: death due to TBI, loss of consciousness for 30 min or more, post-traumatic anterograde amnesia for 24 h or more, a worsening Glasgow Coma Scale full score of <13 in the first 24 h [unless invalidated upon review (e.g., attributable to intoxication, sedation, systemic shock)], and one or more of the following present: an intracerebral hematoma, a subdural hematoma, an epidural hematoma, a cerebral contusion, a hemorrhagic contusion, dura penetration of the TBI, subarachnoid hemorrhage, and brain stem injury (15–17).

The most common neurological scales used for neurological examinations include the Glasgow Coma Scale (GCS) and the Glasgow Coma Scale—Pupils (GCS-P). The Glasgow Coma Scale (GCS) is used to assess the extent of the best motor response, verbal response, and eye opening, and it allows for tentative determinations of TBI prognoses. Depending on the GCS score, the course of traumatic brain injury can be classified as mild (a GCS score of 13–15), moderate (a GSC score of 9–12), or severe (a GCS score of 3–8). The long-term neurological state is determined using the Glasgow Outcome Scale (GOS) and Glasgow Outcome Scale—Extended (GOSE) (15–17).

Primary brain injuries are classified with the qualitative Marshall scale and the Rotterdam CT score. The Marshall scale involves six categories of brain injury depending on computed tomography (CT) images: Category 1 includes cases wherein there are no intracranial pathologies in the CT image; Category 2 includes cases that have a midline shift of <5 mm; Category 3 describes cases of compressed or effaced basal cisterns with a midline shift of <5 mm; Category 4 indicates diffuse injuries that involve a midline shift of >5 mm; and Categories 5 and 6 pertain to cases of surgical mass evacuation, where Category 5 applies to every surgically treated injury, and Category 6 describes injuries of high or mixed density (over 25 mL) that have not been surgically treated (17, 18). The Rotterdam CT score is a classification that has been described relatively recently. The scale is used to assess CT images, pathological lesions, blood in the ventricular system, and subarachnoid bleeding using categories ranging from 0 to 6. Categories 5 and 6 include unfavorable prognoses for patients with TBI (19, 20).

Radiological imaging tests are crucial for determining the nature of brain injuries, their locations, and the indications for surgical treatment. However, many publications have stated that repeated CT scans of the brain are unnecessary in approximately 35% of cases, especially among patients without deterioration in their neurological condition and in cases of mild brain injuries and axonal damage. Imaging data are only one element of urgent diagnostics; thus, they are not sufficient for understanding the mechanism of the injury and cannot serve as the basis for long-term prognoses. Furthermore, the analysis of biological material in patients with TBI in Intensive Care Units (ICUs) has recently garnered significant interest, as it reduces the risks typically associated with transporting intensive care patients to radiology departments. However, biomarkers offer a more comprehensive approach via the analysis of the levels of specific biomarkers to gain a more complete understanding of the disruption of the neural structure integrity, the regenerative capacity of the neural tissue, and the reconstruction and myelination of nerve fibers, neurons, and astrocytes. Therefore, the evaluation of TBI biomarkers complements radiological imaging techniques with the assessment of possible future neurological deficits, thereby allowing for the appropriate course of therapy to be determined and the selection of the optimal, patient-specific rehabilitation, offering the chance to achieve better results (21).

Brain injury biomarker determination is an important supplement to classification and imaging methods. Although there are qualitative methods for assessing brain injuries, the determination of the biomarker concentration is an example of a quantitative method. The performance of biomarker diagnostics has been suggested as a method for quantitatively approaching the issue of prognosis and directly determining the pathophysiological mechanisms of the injury (2). The most frequently used biological materials are serum and cerebrospinal fluid (CSF). Both methods are characterized by invasive entry and the risk of infection. The concentrations of biomarkers are used to diagnose the acute and long-term complications of brain injuries and assess their cellular and molecular mechanisms, thereby improving prognoses and clinical assessments. Additionally, the repeatability of the tests and the short time required to complete laboratory determinations add to the usefulness of these methods (21, 22).

Studies over the last two decades indicate a significantly increased interest in the topic of biomarkers in TBI, with the most common publications presenting nerve tissue protein, cytokine, and coagulation tests. Moreover, TBI biomarkers in different biofluids have also been discovered (10, 23–25).

The aims of this review were to analyze the current state of the knowledge and describe the available biomarkers of traumatic brain injury and their correlation with the stages of brain injury and the clinical prognosis. The literature search was conducted up to 1 July 2024. The PubMed and Cochrane databases were used to identify studies published in English that focused on TBI epidemiology and pathophysiology and the biochemical analysis of TBI biomarkers. The literature search revealed 351 articles from the PubMed database and 41 articles from the Cochrane database. Only international publications written in English were selected. A total of 271 items were excluded, including articles and abstracts. This review included selected literature reviews, as well as observational, experimental, and clinical studies published between 2010 and 2024.

Brain tissue is a complex collection of neuronal cells and accessory elements that are isolated via the blood–brain barrier (BBB) (26). The pathophysiology of traumatic brain injury results from a disruption in the integrity of neural structures [i.e., neuronal bodies and nerve fibers (axons)], as well as from disruption in the cytoskeleton (microtubules and microfilaments) and structure-providing elements (27–29). Mechanical injury causes biological (disintegrative) injury, which then leads to dynamic biochemical changes. Blood–brain barrier leakage results in the secretion of stored and newly synthesized mediators into systemic circulation, allowing for the determination of brain tissue injury mediators in sera. Local biochemical mediators are augmented by systemic mediators and assume the form of the multidirectional and dynamic effects of oxidative stress, oxygen free radicals, interleukins, and apoptotic factors (30–32). The following mediators have been specified: mediators related to the biochemical injury of astroglial cells, neuronal damage demyelination processes, and axonal injury neurodegenerative processes and cytokines (31–37).

The cytokine group is characterized by a complex pleiotropic mechanism of the induction and regulation of local and systemic inflammation. Cytokines are produced locally in elements of brain tissue, as well as via peripheral immune cells, which disturb the assessment of local processes. Immediate gene expression, a rapid increase in the cytokine concentration in body fluids and brain tissue, and a short half-life indicate the favorable properties of these substances as biomarkers. However, their systemic origin and impact are emphasized by their low specificity in relation to brain tissue (21). Pro-inflammatory interleukins augment adverse changes and cell apoptosis, thereby increasing apoptosis protein transcription and intensifying oxidative stress (37, 77). IL-6 and IL-8 determinations are the most widely used, and they reach their peak levels in brain cells on day one after injury and are then considerably elevated on days three to five following the stimulus (78). Under normal conditions, the serum levels are at most 1.8 pg./mL for IL-6 and at most 14.6 pg./mL for IL-8. In TBI cases, the levels are 1,100 pg./mL and 0–2,400 pg./mL, respectively (79, 80). IL-10 is important in the pathogenesis of post-traumatic changes and the regeneration of nervous brain tissue. Local anti-inflammatory action seals the blood–brain barrier and promotes the reconstruction and myelination of nerve fibers, neurons, and astrocytes. The involvement of IL-10 in neuroprotection, neurogenesis, and the regulation of the stress response and hippocampal synaptic plasticity connected with learning and memory has been suggested. Moreover, markers of oxidative stress and antioxidative capacity have been presented as crucial in the pathomechanism of brain tissue injury. The studies demonstrate the association of increased values of specific interleukins (e.g., IL-6, IL-8, IL-10) with increased protein indices (S100B, NSE, GFAP) specific to nervous tissue. Conversely, the literature on cortisol and other specific TBI biomarkers is relatively limited (81–85).

A similar application in TBI prognoses has been described for cortisol measurements. The dynamics of the changes in the cortisol concentrations in blood serum results from the low specificity of this substance toward nervous tissue. Concentration disturbances reflect the hormonal state of the hypothalamic–pituitary–adrenal axis and are also a factor of systemic stress. The description of the bioavailability is noteworthy—determinations in saliva and 24-h urine collections have diagnostic value comparable to that of determinations in blood serum. Tumor necrosis factor (TNF)-alpha assays have little clinical significance due to their systemic origin and effects. In addition, coagulation tests [i.e., the prothrombin time (PT)/International Normalized Ratio (INR)] and the platelet count in peripheral blood counts show that the hemostasis state is not a specific enough biomarker in relation to TBI. The general usefulness of these indices for predicting systemic prothrombotic complications or progressive hemorrhagic changes has been indicated (21). The basic characteristics of the clinical utility of the specific and non-specific biomarkers are presented in Tables 1, 2.

Non-coding, single-stranded RNA molecules with a length of 21–23 nucleotides act as post-transcriptional regulators of gene expression and are involved in mRNA degradation and repression. MicroRNAs, non-coding, single-stranded RNA molecules, have been expressed in the cerebellum, hippocampus, midbrain, and frontal cortex (86) and play a significant role in synapse formation, protein expression, and neuronal network construction (86–88). The significant diagnostic and prognostic value of miRNAs has been described in blood serum and cerebrospinal fluid for MiR-16, MiR92a, and MiR-765 (89, 90). Other miRNA molecules (miR-142-3p, miR-423-3p, miR-425-5p) enable the identification of patients with mild TBI who are at risk of post-brain trauma syndrome and indicate the significant diagnostic and prognostic role of neurotrauma within 48–72 h of the injury (91, 92).

Moreover, the evaluation of salivary miRNAs has shown potential in TBI diagnosis and prognosis assessment. The salivary miRNAs (miR-182-5p, miR-221-3p, mir-26b-5p, miR-320c, miR-29c-3p, and miR-30e-5p) indicate a functional relationship with neuronal development and describe persistent dysregulation for up to 2 weeks after injury (88, 93). miR-27a-5p regulates the sensitivity of neurons to apoptosis and is significant for the protection of the blood–brain barrier (94, 95). The miR-320c values change in both the blood and saliva in severe and mild traumatic brain injury, and the cerebrospinal fluid concentration corresponds with the cerebral cortex (96). Compared to protein markers of traumatic brain injury, miRNAs achieve higher sensitivity due to their stability in peripheral fluids, their ability to cross the blood–brain barrier, and their protection via RNA-binding proteins and exosomes (10, 97).

Biomarker determination in CSF and brain tissue certainly has the greatest medical value. The microdialysis method requires the installation of a special sensor in the brain tissue, as well as other additional bedside devices that enable the reading of the parameters. Unfortunately, this is not an available diagnostic method in every situation. For this reason, the determination of biomarkers in other body fluids is of considerable scientific interest (21).

Cerebrospinal fluid forms the natural brain space and is highly recommended for determining the presence of biomarkers that are initially released from the brain tissue [i.e., S100B, GFAP, UCH-L1, and neurofilaments (NFLs)]. The CSF is an indicator of the tightness of the blood–brain barrier (BBB), as BBB damage is indicated when the serum albumin ratio and the CSF/serum ratio for albumin are increased (10, 23). Because the BBB is disturbed and different mediators (cytokines, cortisol, and TNF) passively penetrate into the CSF and reach higher concentrations than those in the serum, it cannot be understood as a completely true indication of the local intensity of inflammatory processes (23).

Due to the indicated limitations, studies have evaluated the biomarker concentrations in other biological fluids and tissues, such as in whole blood, brain tissue, saliva, urine, and gastric mucosa and via bronchial lavage and oral swabs (21, 24). Most biomarkers in blood have incredibly trace concentrations compared to those in CSF due to the larger distribution volume, the larger intravascular and extracellular water volume, proteolytic degradation, hepatic or renal metabolism, and elimination. For these reasons, there are limitations to the use of blood/serum determinations (23). A small number of experimental studies have presented biomarkers in other body fluids, such as saliva and urine, as well as in fatal cases, which are based on autopsy tests (25, 98). This aspect is the least known and is a space for enriching our knowledge (17, 25, 98–103).

Saliva and urine analyses of patients with TBI may be a real source of molecular biomarkers (104, 105). Janigro et al. (106) presented the clinical utility of assessing the S100B protein levels from the saliva of patients with TBI. In their analysis, the S100B levels were practically four times higher in the saliva samples than those in the control blood samples of patients with suspected TBI. Monroe et al. (107) described a direct correlation between the values of the filament protein light chains (NF-L) obtained from saliva samples and the risk of axonal damage among athletes. Moreover, based on an analysis of saliva samples, Hicks et al. (87) demonstrated that six miRNAs significantly change in patients with traumatic brain injury and thus suggest their use as non-invasive biomarkers of TBI. Through an analysis of the urine samples of patients with traumatic brain injury, Rodriguez et al. (108) showed differences in the S100B kinetic patterns when compared to the blood results. In both analyzed cases, the peak values were reached within 6 h of the injury, and the S100B concentration in the urine decreased gradually by 48 h (but lasted 96 h in the serum) (109). The tau protein has been detected in urine in post-mortem examinations, indicating that this result is a predictive factor of axonal injury, which may be an auxiliary tool for future research and the diagnoses of TBI cases (110–113).

The details of the methodologies and significant technological development have increased the sensitivity and specificity of tests. The mainstream assays used to quantify biomarker levels are enzyme-linked immunosorbent assays (ELISA), polymerase chain reactions (PCRs), mass spectrometry, chromatography, and electrochemical methods. Nevertheless, these methods do not provide the accurate detection of biomarkers at low concentrations. To address this challenge, ultrasensitive methods have been developed, such as digital PCR, rolling-circle amplification (RCA) for the detection of nucleic acids, and meso-scale discovery (MSD) based on electrochemiluminescence technology (114). In 2010, Walt et al. (113, 114) developed a highly sensitive array sensing technology called the single-molecule array (SIMOA). Based on high-density, fiber-optic arrays, the SIMOA allows for the determination of molecules at the single-molecule level, making it a pivotal tool for single-molecule research. Conventional immunoassays, including enzyme-linked immunosorbent assays (ELISAs), chemiluminescence assays, and electrochemiluminescence assays, present low sensitivity levels of approximately 10–13 M (~0.1 pM), while plasma mass spectrometry limits the assay accuracy and is unable to provide quantitative responses. A more widely used technique is immuno-PCR, which increases the sensitivity via the labeling of a detection antibody with a DNA molecule, which is then amplified and quantified via PCR. The sensitivity of these tests has increased 10- to 100-fold over that of conventional immunoassays. An advance in automated immunoassays is the SIMOA analyzer, which integrates the single-molecule sensitivity and multiplexing capability with high-throughput ELISA reagent automation to create an instrumental system capable of molecular-level analysis.

The average sensitivity improvement in SIMOA immunoassays versus conventional ELISAs is more than 1,200-fold, with coefficients of variation of <10% with the limit of detection in fg/mL. The technical challenges of detecting microRNAs in biological fluids are related to the availability of specialized laboratory equipment and appropriately qualified laboratory and medical personnel to perform the sampling. In addition, research on miRNA expression and its diagnostic and prognostic value in TBI is still under scientific evaluation and does not represent a defined standard for evaluation and interpretation. In clinical practice, these factors are the fundamental basis for microRNA analysis (71, 114).

Complex diagnosis schemes based on radiological scans and clinical observations are the main methods for predicting the risk factors of mortality and the neurological state. Various biomarkers have been presented as acute- and chronic-phase TBI mediators that indicate the degree of the brain injury and the neurological condition in critical care (Figure 1). The crucial advantages and clinical prognosis basic on TBI biomarkers are presented in Table 3.

Based on the available publications, it can be stated that there is no definitive and accurate single marker with a high prognostic value for neurological damage to brain tissue; however, the combination of several substances significantly increases the diagnostic value. This approach allows for a holistic assessment of the brain injury stages and for predictions of any complications. Serum and CSF biomarkers are a promising prognostic and diagnostic tool for TBI. There are no additional regulations on the collection and use of other biological materials for diagnostics. Future trends in studies on the markers of brain tissue damage are concentrated on the creation of the optimal brain injury biomarker panel that considers genetic analyses, specific mRNAs, and the use of other biofluids in laboratory examinations when using ultra-sensitive technology, which will allow for the application of the appropriate therapies, thereby reducing the number of complications and the risk of death in patients diagnosed with traumatic brain injury.