Creatine Monohydrate for Mild Traumatic Brain Injury: a Randomised Controlled Pilot Study

INTRODUCTION Mild traumatic brain injury (mTBI), often synonymous with the term 'concussion',1 affects approximately 0.6%-1.2% of the global population each year. Among these, 10-30% experience persistent post-concussive symptoms (PPCS), including headaches, difficulties with concentration and memory, fatigue, sleep disturbances, dizziness, irritability, and anxiety. PPCS not only affects the individual's health but also results in considerable socioeconomic challenges. Research has indicated a 4.2% decrease in the earnings of concussion patients five years after the injury and an increased risk of unemployment. Given the lack of universally effective treatments for PPCS, researchers are exploring alternative approaches such as nutritional supplementation.

One supplement under investigation is creatine monohydrate (CRM), a well-known ergogenic aid among athletes. CRM is considered a potential intervention for PPCS owing to its presence in the brain and its established role in muscle development. The metabolic changes that occur in the brain after an mTBI (e.g. disrupted energy homeostasis and impaired mitochondrial function), suggest that CRM supplementation may offer benefits by reducing PPCS. Although evidence of brain creatine alterations in the acute phase is limited, animal studies suggest concussion-related pathophysiological changes which indicate that creatine supplementation may reduce the severity of concussions. The presence of brain-specific isoforms of creatine kinase, which is essential for the ATP-phosphocreatine system further supports the role of creatine in the central nervous system's energy metabolism. Research has shown that creatine supplementation can enhance cellular energy availability and raise brain phosphocreatine content by up to 15%, thereby improving brain metabolism.

In addition, CRM has demonstrated strong anti-inflammatory properties, such as reducing cytotoxic effects in cells exposed to oxidative stress and inhibiting the formation of reactive oxygen species-induced mitochondrial permeability transition pores. This is relevant given the connection between concussion-induced neuroinflammation and symptomatology. By addressing neuroinflammation, CRM supplementation may contribute to symptom relief. Moreover, CRM may enhance several cognitive and psychological domains that are often compromised in PPCS, such as fatigue, working memory, and mood state. Creatine has also been associated with reductions in chronic fatigue, depressive symptoms, and anxiety, further indicating its potential for managing PPCS. Recent findings indicate possible benefits in stressed populations, which may be relevant for individuals experiencing PPCS after concussion.

The aim of this study was to examine the effect of CRM on PPCS, as assessed using the Rivermead Post Concussion Symptoms Questionnaire (RPQ), and to explore the feasibility of this pilot study.

METHOD Patient and public involvement No patient or public involvement was undertaken at this stage although public input informed a protocol amendment extending the concussion timeframe from 6-12 to 6-18 months. For the planned larger RCT, patients will contribute to study design, recruitment processes, and the selection of outcome measures.

Trial design A pilot RCT with three arms (intervention (INT), placebo (PLA), and control (CON) groups). The study was retrospectively registered in a study protocol, at ClinicalTrials.gov (NCT06644131) on 30 September 2022. The record was updated on 21 January 2026. The study was conducted in accordance with the CONSORT 2010 guidelines for reporting a pilot or a feasibility trial.

The investigators initially aimed to recruit 45 participants through outpatient clinics, social media platforms (Facebook, Instagram, LinkedIn, and X), and newspaper advertisements. Ultimately, 34 participants were enrolled, all of whom completed the study. Participants were enrolled between September 2023 and January 2024. The trial was conducted in participants' homes. Self-report questionnaires were administered to all three groups at three time points: upon enrolment, at week 4, and at week 8 (end of testing period).

Intervention Participants in both the INT and PLA groups were instructed to consume a powdered supplement daily. The PLA group received a powder that visually resembled CRM but had no discernible nutritional value. Both groups followed the same dosing protocol, consuming 5 g of the powder per day for seven weeks. In both groups, the powder was formulated into tablets and administered in equivalent doses.

Unlike previous studies that employed a loading phase of 0.3 g/kg/day during the first week when examining the effects of CRM on muscle function, investigators opted for a fixed daily dose of 5 g throughout the study period. This decision was made to improve participant adherence and minimise the risk of gastrointestinal discomfort commonly associated with high-dose creatine intake.

Statistical analyses The trial was powered at 80% (α = 0.05) to detect an effect size of 0.4 in RPQ score, which the investigators considered adequate to determine clinically meaningful differences between interventions.

Normality was assessed using Q-Q plots for all variables to ensure appropriate test selection. Some variables were analysed using non-parametric tests despite meeting normality assumptions, with the aim of providing the best basis for comparison. This decision was made in consultation with a statistician. When comparing baseline values between groups, the investigators used either Kruskal-Wallis or ANOVA. For within-group analyses from pre to post, the investigators used either the Wilcoxon signed-rank test or the paired t test.

The investigators report the summarised scores for RPQ-3, RPQ-13, and total RPQ rather than individual items, as this approach yields more accurate results. Data on each item are available in the supplemental material (Tables S1 and S2). This decision was made in consultation with a statistician.

Analyses followed the intention-to-treat principle, with participants assessed in their original randomisation groups regardless of adherence. All statistical analyses were performed using IBM SPSS Statistics for Mac (version 30.0.0.0 (172)).

Harms Within the dosage range used in this study, CRM is not associated with expected risks or adverse side effects, even with use exceeding one year, apart from potential weight gain resulting from increased muscle mass.

Sample size The study aimed to recruit 45 participants through a convenience sampling method; however, because of recruitment challenges, the final sample consisted of 34 participants. Given the limited number of RCTs available for informing power calculations and the absence of a universally accepted clinically relevant change in score, this pilot study was also conducted to inform sample size estimation for a future RCT.

Randomisation and blinding Randomisation was conducted in advance using an online tool (http://www.jerrydallal.com/random/random_permutation.htm). Both the INT and PLA groups were blinded to the type of powder they received. To maintain blinding, participants were assigned unique identification numbers. A researcher not otherwise associated with this study was responsible for the randomisation process. They also managed the distribution of the powders and provided the accompanying instructions to participants.

Ethical considerations The pilot study was approved by the National Committee on Health Research Ethics (S-20230071) and the Danish Data Protection Agency (11.651). The study was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki. Prior to participation, all patients received verbal and written information about the study, and they provided written informed consent. Participants were informed that they could withdraw at any time without consequence. Data from withdrawn participants were excluded from the analysis. Data were entered into the SurveyXact Online Clinical Trial Management System. To ensure anonymity, participants received a registration number as soon as they were deemed eligible.

RESULTS A total of 34 participants were enrolled in the study, of whom 88% were women. Participants had a mean (SD) age of 34.24 (5.90) years, body mass of 72.44 (13.56) kg, and height of 171.53 (6.55) cm. During the study period, 10 participants were randomised to the INT group, 10 to the PLA group, and 14 to the CON group. All participants completed the follow-up period, and no attrition occurred after allocation to the three study groups. This indicates a high level of usability and practicality.

Participants' baseline characteristics were comparable across the three groups.

The changes from baseline to follow-up at week 8. No significant differences were observed between the groups. There were no significant differences for the total RPQ score (p = 0.21), RPQ-3 (p = 0.49), or RPQ-13 (p = 0.18).

The estimated effect sizes for group differences in total RPQ scores and the RPQ-3 and RPQ-13 subscales. A moderate effect size was observed for the total RPQ score (-0.46) and for RPQ-13 (-0.47) when comparing INT with CON. A small effect size was found for RPQ-3. Similar patterns were observed when comparing PLA with CON, with effect sizes of -0.51 for the total RPQ score, -0.57 for RPQ-13, and -0.18 for RPQ-3.

No adverse events were reported throughout the study.

DISCUSSION Summary of findings All participants completed follow-up, and no attrition occurred after allocation to the three study groups. This indicates that the the study demonstrated strong usability and practical feasibility. No baseline differences in demographics, training hours, or RPQ scores were observed between groups. After eight weeks, there were no significant between-group changes in weight, training hours, or RPQ scores. Effect size estimates indicated moderate differences between INT and CON for RPQ-13 and total RPQ scores (Cohen's d = -0.47 and -0.46) and a small difference for RPQ-3 (-0.22), although none reached statistical significance. Similar effect sizes were observed between PLA and CON.

Interpretation To our knowledge, few intervention studies have examined the effects of creatine supplementation on traumatic brain injury or post-concussive symptoms. This study is the first to assess CRM supplementation in adults with PPCS. The investigators hypothesised that CRM would reduce symptom severity, as measured by the self-report RPQ.

Two studies by Sakellaris et al. in children and adolescents with moderate to severe TBI reported significant functional and neurological improvements, including reduced post-traumatic amnesia; shorter hospital stays; and fewer symptoms such as headache, dizziness, and fatigue. A study by Turner et al. found enhanced cognitive function and increased corticomotor excitability in healthy adults under hypoxic conditions, indicating CRM's potential as a neuroprotective intervention. Other studies have investigated CRM supplementation in relation to PPCS. Brooks et al. reported no effect on depression, anxiety, or stress following a six-week protocol using 0.1 g CRM/kg body weight. In contrast, Sherpa et al. suggested that CRM may have potential as an adjunct to cognitive-behavioural therapy for depression.

Despite these promising trends, the evidence is limited by small sample sizes and variability in study quality, which warrants caution in interpretation. Nevertheless, these findings contribute to an emerging field that may inform the development of future clinical interventions for PPCS.

Strengths and limitations One strength of this study is its randomised controlled design, which reduces potential sources of bias. Additionally, the use of the RPQ enhances reliability, as it is a widely used and validated tool in concussion research. All analyses were conducted according to the intention-to-treat principle, with imputation applied for missing data. The study groups were comparable at baseline, and no participants were lost to follow-up, indicating a feasible and robust study design. Recruiting participants from both clinical settings and social media platforms enhanced the study's ecological validity by increasing sample diversity.

Nevertheless, several limitations should be noted. First, the relatively small sample size limits the statistical power, reducing its sensitivity to detect subtle but clinically meaningful effects. It is worth noting, however, that the sample size is comparable to that of other studies in the field. Second, although blinding was maintained, participants' expectations regarding CRM may have influenced self-reported outcomes. Third, while adherence to the supplementation protocol was monitored, dietary creatine intake and other lifestyle factors that may affect creatine metabolism were not controlled for. This approach reflects common practice in CRM supplementation research, which supports comparability across studies. Nonetheless, more stringent control of these variables would likely have strengthened the reliability and interpretability of the findings.

Implications for future research The short duration of this study and the absence of magnetic resonance spectroscopy (MRS) limit the conclusions that can be drawn. Slankamenac et al. and Stockler-Ipsiroglu reported that extended supplementation (six months or longer) can increase brain creatine and alleviate symptoms, highlighting the need for longer PPCS trials with baseline and follow-up MRS assessments. Future trials should incorporate larger samples, longer follow-up periods, and optimised dosing to clarify CRM efficacy. Moreover, incorporating MRS could provide objective evidence of changes in brain creatine and strengthen mechanistic interpretation, as demonstrated in several previous studies. A larger randomised study, ideally with MRS, would also allow detailed analysis of RPQ subcomponents to determine whether any observed effects predominantly reflect physical symptoms, such as headache, or neurocognitive symptoms, such as impaired focus.

Because of recruitment challenges in this pilot study, the inclusion criteria in the planned RCT will be broadened. The revised criteria are as follows:

• The allowable time since concussion will be extended from 6-18 months to 1-60 months.
• The eligible age range will be expanded from 20-45 years to 18-65 years.

Additionally, one exclusion criterion will be eliminated:

• Elite athletes and people who are physically active for more than 10 hours per week on average.

CONCLUSION The study showed strong usability of assessment tools and practical study procedures, supporting the protocol's suitability for larger trials evaluating CRM for PPCS. Reported effect sizes are exploratory and not evidence of efficacy. Larger randomized controlled trials are needed to clarify CRM's potential as a safe, accessible adjunctive treatment for PPCS.