Levels of biomarkers associated with subconcussive head hits in mixed martial arts fighters

Participants

This exploratory study enrolled 40 participants aged between 18 and 32 years (25.4 ± 3.8) carried out between May and August 2019. The participants were divided into three groups: (1) MMA fighters; (2) Active control group: active individuals according to the International Physical Activity Questionnaire –IPAQ, an instrument for measuring the physical activity level (IPAQ Research Committee, 2005), and more precisely muscle training practitioners; (3) Sedentary control group: healthy sedentary individuals according to the Brazilian version of the Sedentary Behavior Questionnaire (Mielke GI, Owen & PC, 2014). Thus, a total of 16 MMA fighters from three training centers of a city in Northeast Brazil compared with 14 resistance training practitioners, and 11 healthy sedentary men recruited through social media (Fig. 1).

Ethics

Study procedures were approved in December, 2017 by the Research Ethics Committee of the Federal University of Rio Grande do Norte (protocol reference #3.246.228), and all participants provided written informed consent prior to the study.

Procedure

A member of the research team (graduation and post-graduation students) provided information about the study and requested consent to drain blood samples and use the Sports Concussion Assessment Tool 5th edition (SCAT5) results for research purposes.

For the group of fighters, the collections were carried out in a controlled environment, during the sparing session, which is fight training involving two individuals under the supervision of an experienced trainer or technician. For the present data collection, only one session was carried out, lasting 3-4 rounds of 5 min each. As for the equipment used during the fight, the participants only wore gloves specific to MMA, with protection for the wrist and finger joints.

The individuals of the control group (Healthy Active Group) were instructed not to exercise on the data collection day for all the following tests, and the fighters were instructed not to participate in any contact training during the study period. The group of healthy sedentary individuals was instructed to maintain their normal routine of daily activities. The assessments (SCAT5 and blood collection) were carried out at three moments: immediately before the fight, up to 30 min after and 72 h after the fight for fighters and at a single moment for active and sedentary individuals.

Covariates

Sociodemographic and anthropometric data of the individuals (age, education level height, weight, and body index mass (BMI), lifestyle habits (smoking history and alcohol intake), were evaluated by self-report, and physical activity frequency were collected in the baseline assessment.

Concussion symptoms evaluation

Sports Concussion Assessment Tool 5 (SCAT5): the SCAT5 is a standardized questionnaire to evaluate athletes with concussion. It can be used to evaluate children from 13 years old (Patricios et al., 2023). The questionnaire consists of 6 evaluation steps; however, this study only used the first part of the tool regarding the evaluation of the number and severity of the symptoms, according to the studies of Meier et al. (2017a) and Di Battista et al. (2018). The translation of the symptoms scale performed and used by the Brazilian Football Confederation was used for this purpose. The SCAT5 was applied three times (see Fig. 2).

The symptoms score is composed of 22-item symptoms using a seven-point Likert scale rating. The severity of the symptoms is obtained by the sum of the rated symptoms score (Guskiewicz et al., 2013). This scale has shown to be reliable for assessing symptom presence and severity (Luoto et al., 2019).

Blood sample collection

Eight milliliters of venous blood were drawn from the participants by a certified nurse using EDTA K 2+ tubes. The blood collection was done at three time points. After a short cooling down period, the samples were stored in a cooler with ice at a temperature of around 17 °C. The samples were centrifuged shortly after in a centrifugal machine (CentriBio 80-2B) at about 2,500 revolutions per minute for 10 min in order to separate the blood plasma. The plasma was stored in microtubes in aliquots of 500 µl and maintained at a temperature of −80 °C until the sample analysis. Blood samples were all collected at the same time of the day, in the afternoon, and none of the participants were fasting. At the time of the first blood collection, for all fighters was requested to avoid any head impacts or head trauma for 5 days prior, following the researchers’ recommendations.

Biomarker analysis

Plasma concentration of three blood biomarkers (UCH-L1, GFAP and BDNF) were determined using the Enzyme-linked Immunosorbent Assay (ELISA) from Elabscience (Houston, TX). First, duplicate samples of 100 µl of each subject were used following the manufacturer’s instructions. The reading was performed at 450 normetanephrine (nm) in an ELISA reader (Mindray MR-96A) and the results were interpolated into a standard curve for each marker.

Statistical analysis

Quantitative variables are presented for central tendency (mean) and dispersion measures (standard deviation), while the categorical variables were expressed for absolute and relative frequencies. The Shapiro–Wilk test was utilized to check the data normality. The difference between groups related to the categorical variables was assessed by the Qui-square test (for descriptive analysis). A one-way ANOVA with Tukey’s post hoc was performed to compare the biomarker levels between fighter group and healthy active and sedentary groups in the initial evaluation. After that, ANOVA for repeated measures was performed to analysis of the difference between the pre-sparring moment, immediately after and 72 h after the sparring session, the number and severity of symptoms and levels of concussion-related biomarkers (only for the fighter group). Effect size (Cohen’s d) was calculated for biomarkers values between groups (in baseline). Values between 0.20 and 0.49 were considered small effect size, 0.50 to 0.79 represented moderate size, and over 0.79 was considered as a large size (Cohen, 1988).

All data were analyzed using the GraphPad Prism version 8.2 software program (GraphPad Inc., La Jolla, CA, USA). A statistical significance of 5% was established for all analyzes.

Sample characterization

A total of 30 male individuals were included in this study after the selection and meeting the inclusion criteria (10 fighters, 10 resistance training practitioners, and 10 healthy sedentary men). Participant characteristics, socio-demographic data, and concussion symptoms and severity are listed in Table 1. There was no difference between the groups related to weight, height, and BMI. The fighters had a mean of 6.4 ± 3.2 years’ experience of professional practice with a frequency of 13.4 ± 0.8 h training sessions per week (p < 0.0001). In the variables related to concussion, evaluated through the SCAT5 at the three moments, the number of symptoms before the fight was higher in the group of fighters (SCAT5 - Number of Symptoms = 6) when compared to both control groups (SCAT5 - Number of Symptoms = 0.9 and 4.1 for Active control and sedentary group, respectively) (p = 0.01). The fighter’s group had a greater severity of symptoms before the fight (SCAT5 = 11.3) compared to the active healthy group (SCAT5 = 1.3) and sedentary healthy group (SCAT5 =7.2), with a statistically significant difference (p < 0.01).

When comparing the number and severity of symptoms, as measured by the SCAT-5, only in the group of athletes across the three assessment times, no significant differences were found (p-value: 0.73 for number of symptoms and p: 0.86 for severity of symptoms).

Biomarker analyzes

In assessing BDNF, it was possible to identify differences between the fighters and active control in baseline (p = 0.03) (Fig. 3A). There were no significant differences between groups for UCH-L1 or GFAP markers during the first assessment (Figs. 4A and 5A). It was further possible to identify a significant difference between the immediate and 72 h post sparring moments when evaluating BDNF (p = 0.03) (Fig. 3B). However, for UCH-L1 levels there was a significant difference between pre-sparring and immediately post-sparring (p = 0.04) (Fig. 4B). There were no significant differences between the different assessment times for the GFAP levels (Fig. 5). Differences in BDNF levels were observed between the fighters immediately after the sparring session and the active control group (p = 0.01) (Fig. 3C), however, we did not find a difference between 72 h post-sparring and the control groups (Fig. 3D). No differences were found in UCH-L1 or GFAP levels (Figs. 4C and 4D; 5C and 5D). The levels of the other proteins revealed no differences between groups. The values of each biomarker, according to the groups, are available in Table S1. It is possible to observe in S1 the results of effect size (ES) between biomarker measurements. For BDNF the effect size was large, while ES for UCHL-1 was small. It was not possible to calculate the ES for GFAP.

BDNF

The use of BDNF as a marker of the effects of physical activity and exercise is well documented (Huang et al., 2014). In addition, BDNF is clinically important due to its role in neuronal integrity because it expresses functional status (memory processing) and regenerative power (synaptic plasticity) (Cohen-Cory et al., 2010). An assessment of serum levels can be an important diagnostic tool and help to clinically manage alleged subconcussive scenarios of individuals exposed to trauma.

High BDNF levels were observed in the fighting group when compared to the active group in the initial assessment in our study (588.5 (79.0) vs 423.9 (144.3), respectively), like Oztasyonar (Oztasyonar, 2017) who found significantly higher post-training BDNF levels in the taekwondo fighters, boxers, and athletes groups compared to pre-training level. These findings may be supported by the type and level of physical activity intensity practiced by the groups (Oztasyonar, 2017). The active group practiced strength exercises; on the other hand, the group of fighters developed a training routine with an average frequency of 12 times a week, predominantly consisting of aerobic exercises such as running and cycling.

Earlier studies have shown that basal BDNF levels are low in people who do not exercise or participate in sports. BDNF levels seem to have an important relationship with regular exercise, which has been proven to have positive effects on angiogenesis, neurogenesis, and brain functions (Nofuji et al., 2012; Damirchi et al., 2014; Manley et al., 2010; Yue et al., 2013). Moreover, despite being exercise-induced, serum BDNF quantification corresponds to cerebral production, since muscle-produced BDNF does not appear to be released into the bloodstream (Matthews VB et al., 2009).

Our results show that BDNF maintained similar levels to baseline after the sparring, which is like one study with Olympic boxers (Neselius et al., 2013), but differs from another study which showed elevated serum BDNF concentrations in practicing combat sports at a high level (Ziemba et al., 2020). One possible reason the elevation in this protein was not identified in our study would be the time taken to collect the samples, since blood collection from our subjects occurred very early immediately after the sparring session. According to Manley et al. (2010) and Yue et al. (2013), it is important to collect biomarker samples at least 24 h after brain injury; however, it cannot yet be stated that early assessments can directly influence the results, since other studies present different assessment moments in their methodology (Korley et al., 2016b; Oztasyonar, 2017). Therefore, a consensus is necessary to clarify the appropriate moment for this assessment.

Another explanation for the difference is the type of sport practiced. While our participants engage in MMA, in the study by Ziemba et al. (2020), athletes were divided into karate, taekwondo, and a combination of judo, wrestling, and sumo, which involve different forces of attack, defense, and counterattack, directly related to the aggressiveness involved in sports practice and consequently, with the blows given and received to the head. Additionally, the athletes analyzed in this study are high-level sports people, who present a high training frequency compared to the participants in our study.

The nature of our outcomes considering their frequency and especially the intensity of the blows may in turn have not been enough to result in an expression of increased levels of this protein. Finally, our results show a significant reduction in BDNF levels between the moments immediately after the sparring session and 72 h after the sparring session. In clinical practice, BDNF levels are lower in patients who have suffered Traumatic Brain Injure (TBI) when compared to controls on the day of injury, constituting a biochemical parameter which signals a diagnosis of TBI (Korley et al., 2016a). Our study evaluated the fighters immediately after the sparring session, and therefore we believe that no reduction in BDNF levels was found at this time due to the time elapsed between the end of the sparring session and the immediate collection.

UCH-L1

Our results showed significant differences in UCH-L1 levels in athletes before the sparring and immediately after the sparring session, with a decrease of this marker being observed. However, these findings are divergent as most studies show an increase in UCH-L1 levels after concussion (Brophy et al., 2011; Papa et al., 2012; Puvenna et al., 2014a), furthermore, the results coincide with other studies (Papa et al., 2019; Papa et al., 2016), which showed that a reduction in UCH-L1 can be detected immediately after trauma and remains low at 72 h after the sparring session. According Mondello et al. (2012) UCH-L1 levels in athletes subjected to head impacts can increase up to 6 h after injury, and although they fall rapidly in the following hours, the levels tend to remain significantly elevated for up to 36 h.

About sub-concussion, the study by Harrell et al. (2022), which evaluated rugby players and showed that individuals in the control group did not have a significantly higher serum concentration of UCHL1 than the recreationally active control group. or the pre-game group. Like these findings, the study by Puvenna et al. (2014b) in football players showed that serum UCHL1 concentration did not correlate with the number of subconcussive hits to the head. This study also agrees with our findings that UCHL1 levels were unable to distinguish between post-game football players, emergency room patients who suffered mild traumatic brain injury and healthy controls.

The ubiquitin-proteasome complex is associated with cellular vitality by regulating homeostasis. Accumulation of unwanted proteins because of ubiquitin-L1 dependent defective proteolysis may contribute to events underlying the pathogenesis of several major human neurodegenerative diseases such as Alzheimer’s and Parkinson’s diseases (Layfield et al., 2001); therefore, successive brain injuries, accumulated over years in the practice of MMA, can bring about important impairments that affect the physical, cognitive and psychosocial function of individuals (Riederer et al., 2011). However, the data from the present study do not support this statement, and longitudinal studies are necessary to reinforce this hypothesis.

GFAP

Our study found no differences between GFAP levels at the three assessment points of the fighters, although a recent study showed that it was possible to detect GFAP within one hour of injury, peaking after 20 h and declining slowly until 72 h after trauma (Papa et al., 2019). We believe that our findings are due to the severity of the brain injury, since GFAP is associated with intense brain trauma (Okonkwo et al., 2013) sub-concussive trauma, as it is milder, may not cause typical signs and symptoms, presenting a very low expression as shown in the study by Papa et al. (2019), which possibly explains the lack of significance in our results.

GFAP is a highly regulated protein whose expression is induced by multiple factors such as brain injury and disease (Eng, Ghirnikar & Lee, 2000). GFAP is a classical marker for astrocytes known to be induced by brain damage or during degeneration of the central nervous system (Middeldorp & Hol, 2011). This protein is a promising biomarker of brain injury in patients with concussion in the acute phase (Yue et al., 2019); in addition, GFAP can differentiate individuals with and without abnormalities in computed tomography (CT), being the strongest predictor among the evaluated biomarkers (Gill et al., 2018). Cerebrospinal fluid GFAP levels were detected as acute and cumulative effects of head trauma, suggesting minor central nervous system injury (Neselius et al., 2012) and plasma GFAP concentrations also correlate with MRI lesion types (Yue et al., 2019). However, in our results we did not find significant differences about GFAP in the acute phase of brain injury.

SCAT5

We found differences between the number of symptoms and the severity of symptoms in the SCAT5 before sparring, immediately after the fight and 72 h after the sparring in the group of fighters. Our findings corroborate the results of Meier et al., who observed significant differences in the SCAT5 in American football players, between the pre-training, six and 24-48 h after the trauma, since during the championship (Meier et al., 2017a). However, unlike what was found in the study by Unnsteinsdottir Kristensen, Kristjánsdóttir & Jónsdóttir (2021), in our athletes, sub-concussive symptoms were greater at time 01, that is, before sparring.

We believe that SCAT-5 result is linked, as already mentioned, to the fact that our outcome variable is a subconcussive event and not a concussion. A concussion occurs when a direct or indirect impact is applied to the head with sufficient intensity to generate symptoms (Belanger & Vanderploeg, 2005). However, in a subconcussion, there is a transfer of mechanical energy to the brain with sufficient intensity to affect neuronal integrity (Bailes et al., 2013), but without resulting in clinical symptoms.

A study of 60 boxers found that most impacts suffered during a 2-minute sparring session were below the threshold for concussion (Stojsih et al., 2010). The same could have happened in our study, since the acceleration/deceleration forces, although the intensity for damage to neurons is not yet known (Montenigro et al., 2015), exert greater influence on the symptoms presented by the fighter. Additionally, the SCAT-5 is useful immediately after injury in differentiating concussed from non-concussed athletes, but its utility appears to decrease significantly 3–5 days after injury. The symptom checklist, however, does demonstrate clinical utility in tracking recovery.

From a clinical point of view, we found that MMA practitioners had a greater number and severity of symptoms and severity at baseline than sedentary individuals and active, like what was observed by Petit et al. (2020), who found that athletes participating in contact sport reported a severity of symptoms almost twice as high as non-contact athletes. Therefore, these results indicate that even in healthy conditions (free of continuous concussions), practitioners of contact sports tend to have persistent symptoms, which can later influence cognition and neurological changes.

The results of this study reinforce the need for a discussion on primary concussion prevention, aimed at reducing the risk of recurrent injuries and the potential for persistent symptoms. However, further studies addressing different types of sports are needed. According to the Consensus Statement on Concussion in Sport (Patricios et al., 2023), healthcare professionals should be encouraged to identify and optimize SRC prevention strategies in their environment. Implementing primary SRC prevention at all levels of sport is a priority that can have a significant impact on public health and reduce the risks of compromising athletes’ physical, mental, and cognitive health.

Strengths and limitations

The small sample size in the study may raise limitations regarding the extrapolation of results. However, the subjects’ selection criteria, as well as the calibration of subjective and biochemical measures can minimize potential selection and measurement biases. In addition, it is worth highlighting the pioneering nature of the study in demonstrating the effects of repeated subconcussive blows to the head in MMA fighters. The use of reliable detection methods and duplicate analysis for all markers reduces the risk of possible errors in biochemical measurements.

This study did not evaluate the intensity of the strikes suffered by the fighters, which may be more related to the severity of brain damage than the number of blows. We believe that some aspects of this study could be improved if this assessment were performed. Moreover, this study provides data on the biomarker levels associated with concussion in MMA fighters, as well as the symptoms referred by them. These data are important for understanding traumatic events in the head, especially in MMA fighting athletes who are submitted to these events weekly.

This was a pioneering study which demonstrated that repeated subconcussive strikes in MMA fighters are associated with reduced levels of UCH-L1 immediately after the fight, as well as reduced levels of BDNF 72 h after the fight. Furthermore, the traumatic events were not able to change the levels of GFAP. These exploratory findings may help in understanding the influence of repeated subconcussive blows on MMA fighters, a population frequently affected by these events.