A systematic review was conducted to analyze the role of nonpharmacological management of pediatric concussion. The included studies were identified through a search of studies indexed in the PubMed, Scopus, Web of Science, and Cochrane Database of Systematic Reviews databases between January 2013 and July 2023.The 2020 version of the Preferred Reporting Items for Systematic reviews and Meta-Analyses guidelines was used [13]. The search string used included "mild head injury," "post-concussion syndrome," "pediatric," and non-pharmacological treatment–related terms such as "rest," "physical activity," "vestibular therapy," "vision therapy," "acupuncture," "hyperbaric oxygen therapy protocol," "family intervention," "conventional occupational therapy," "mindfulness and yoga," "collaborative care intervention," "cervical spine rehabilitation," "blue light therapy," and "gaming computer." This study was registered within an Open Science Framework.

The scope of this study was restricted to randomized controlled trials (RCTs) that explored the effect of nonpharmacological treatment on pediatric concussion in an acute or chronic care stages as stated in the Patient, Intervention, Comparison, Outcome, Timing, and Study Design framework (Table 1). "Acute" was defined as nonpharmacological management administered in the first 24-hour following impact. The terms "concussion" and "mild traumatic brain injury" are used interchangeably in this review. All studies that met the following criteria were included: (1) RCT design; (2) focus on children (0–18 years); (3) investigation of the efficacy of nonpharmacological treatment for acute and persistent postconcussive symptoms; and (4) inclusion of eligible outcome measures, including any clinical symptoms and outcome measures involving the acute and chronic treatment periods. We excluded studies of conditions not related to TBI, that were not published in English, for which the full text was unavailable, and that were not included in an RCT. Systematic reviews and meta-analyses were also excluded, but a manual search of the reference lists of the relevant primary studies was conducted.

After removing duplicates, 2 independent authors screened all titles and abstracts of the retrieved records. Two independent authors then retrieved and assessed the full texts of all shortlisted abstracts. All disagreements were resolved through discussions with the full team. The following data were extracted: (1) year of publication, (2) author names, (3) study design, (4) definition of concussion, (5) primary injury cause, (6) participant characteristics, (7) intervention, (8) duration between concussion and intervention, (9) duration of intervention, and (10) primary outcome. The data extraction was conducted by one author and cross-checked by the main author (Fig. 1).

The Cochrane risk of bias tool for randomized trials version 2 (Risk of Bias-2 tool, RoB 2) was used to evaluate the risk of bias of the included studies [14]. Studies were categorized as having a low risk of bias, some concerns, or a high risk of bias based on the 5 domains.

A flowchart of the study is shown in Fig. 1. The initial search yielded 110 titles. A total of 64 records were screened; of them, 18 were excluded and 46 were screened for eligibility. Of the 41 reports assessed for eligibility, only 16 met the inclusion criteria.

We identified 16 RCTs that addressed nonpharmacological interventions (Fig. 2) across pediatric populations (Table 2). The terms "concussion" and "mild traumatic brain injury" were used interchangeably, and concussion-defining criteria differed slightly across studies; however, we adopted the definition offered by the 5th International Conference on Concussion in Sport held in Berlin and endorsed by the American Association of Pediatrics of “a traumatic brain injury induced by biomechanical forces [15]. They further elaborated on 5 criteria: concussion may be caused by a direct blow to the head, face, neck, or elsewhere on the body with an impulsive force transmitted to the head; concussion typically results in the rapid onset of short-lived impairment of neurological function that resolves spontaneously; concussion may result in neuropathological changes, but its acute clinical signs and symptoms largely reflect a functional disturbance rather than a structural injury; concussion involves a lack of visible abnormalities on standard neuroimaging studies; and concussion causes a range of clinical signs and symptoms that may or may not involve a loss of consciousness. The above signs and symptoms cannot be explained by other causes.

Our review further elaborates on concussion symptoms as being both global and specific. Global symptoms can range from a loss of consciousness, fatigue, and drowsiness to other symptoms that are not limited to a single organ system. Specific symptoms were defined as those that could be identified in a single organ system, such as vertigo, hearing impairment, visual difficulties, cognitive problems (difficulty learning), or psychotic symptoms (changes in personality). While unconventional, these symptom definitions are important in this review in the identification of specific nonpharmacological treatments.

Based on this definition, a large number of studies were included in this review. The included studies comprised 1,069 subjects, of whom 535 received nonpharmacological interventions and 534 were enrolled as controls. The most frequent mechanism of injury was sports-related. Three studies were conducted in the acute (<24-hour postinjury) setting, whereas the other 13 were conducted in the chronic or persistent setting (>24-hour postinjury). The study characteristics and key findings of the individual reports are summarized in Table 1.

Postconcussive symptoms were assessed as an outcome measure in nine of 16 studies. Eight studies used the postconcussion symptom scale (PCSS) score as the main outcome [16-23]. Other outcomes included medical clearance to return to sports, severity of depression, quality of life, performance of daily activities, balance, dizziness, and cognitive function [17-26]. Most studies used multiple outcomes, including both patient or parent reports and objective performance measures. The measures were typically assessed pre- and postintervention.

To further enhance the quality of this review, a meta-analysis was performed whenever possible. The meta-analysis was performed using Review Manager ver. 5.4 (RevMan 5, Cochrane Collaboration, Oxford, UK). A heterogeneity test was first performed to reveal data variability using the I² test. I² >50% signifies heterogeneous data; thus, further analysis should employ a random-effects model, whereas I² 50% signifies homogenous data, which allows for a fixed-effects model. Continuous data were analyzed by examining the mean difference, whereas categorical data were analyzed by examining the risk ratio. A funnel plot was created to ensure the absence of publication bias.

Three studies were RCTs conducted in an acute setting (24 hours). Two were evaluated at rest, including prescribed rest and strict rest, while the others were evaluated on permitted screen time and abstinence. The most frequently reported consequence was an acute headache (2 of 2 reports) [20,26].

In an RCT of 93 subjects, Sufrinko et al. [20] showed that rest was associated with improvements in symptoms and verbal memory in patients with signs of injury. Immediate postconcussion assessment and cognitive test and PCSS were used to assess patients on days 3 and 10 for 11–18-year-olds who were tested for 5 days. In an RCT of 99 subjects, Thomas et al. [26] found that enforced rest for 5 days resulted in higher daily postconcussive symptoms and slower symptom recovery than normal treatment. However, the risk of bias associated with this study affects the strength of the evidence. Due to the heterogeneity of the 3 included studies, no meaningful meta-analysis could be performed.

Thirteen reports from 16 RCTs investigated nonpharmacological therapies for chronic or persistent mild TBI. The most frequently evaluated outcomes (13 of 13 reports) were postconcussive symptoms, visual impairment, vestibular problems, depression, and neurocognitive impairment [16-28].

The literature has described nonpharmacological interventions for global symptoms, such as aerobic exercise, collaborative care intervention, cervical spine rehabilitation, and education by physiotherapists [17,23,25-28]. There are also several nonpharmacological treatments for specific symptoms. In essence, patients with acute headache after concussion may benefit from 3 days of rest. Patients with global symptoms should consider aerobic exercise, collaborative care intervention, and cervical spine rehabilitation, while those with specific symptoms may benefit from symptom-based treatments. Patients with cognitive symptoms require hyperbaric oxygen therapy control or family intervention, whereas patients with psychotic symptoms may only benefit from family intervention. Finally, patients with visual symptoms could use 30-minute virtual reality and conventional occupational therapy or virtual rehabilitation, and patients with vestibular symptoms could benefit from vestibular rehabilitation, coordination exercises, and balance exercises. This is further summarized in the flow diagram of the evidence-based findings (Fig. 2).

Chan et al. discovered the safety, tolerability, and potential efficacy of active rehabilitation, including aerobic training, coordination exercises, and visualization and imagery techniques with a physiotherapist among 19 adolescents with persistent postconcussive symptoms lasting >1 month after sports-related injuries compared to controls [16]. Choi et al. [17] found a significant ability of 30 minutees of virtual reality and 30 minutes of conventional occupational therapy to improve unimanual dexterity, activities of daily living, and forearm articular movement among 80 subjects versus controls. Carney et al. [18] found no significant difference between the standard care and family intervention group in terms of pediatric TBI postdischarge care. However, the children with better outcomes lived with their families and had better function at 6-month postinjury.

Kontosetal. [19] reported no significant changes in vestibular rehabilitation or control in 50 subjects. However, while both groups recovered similarly in terms of dizziness, the overall concussion versus behavioral management control group showed a more dramatic clinical improvement in vestibular items.

Hadanny et al. [21] reported that hyperbaric oxygen treatment and sham treatment improved cognitive performance, persistent PCS, behavioral symptoms, and quality of life in children aged 8–15 years. In another RCT, Kurowski et al. [27] reported a greater rate of improvement in the subsymptom exacerbation aerobic training group among 30 subjects versus controls. Bailey et al. [24] found significant improvement in the subsymptom threshold exercise program and control group at 6 weeks after removal of the influence of depression. McCarty et al. [23] showed that, after a collaborative care intervention reported 6 months after the baseline assessment, 13.0% of intervention patients versus 41.7% of controls reported high levels of postconcussive symptoms (P=0.03), and 78% of intervention patients and 45.8% of control patients reported a 50% reduction in depressive symptoms. Leddy et al. [25] reported that after aerobic exercise, children aged 13–18 years recovered faster (13 days) than those assigned to placebo-like stretching (17 days) (P=0.009). Dobney et al. [28] reported the feasibility of active rehabilitation (aerobic exercise, coordination drills, visualization, and education or reassurance) offered early (starting 2-week postinjury). Micay et al. [22] reported that the administration of an aerobic exercise intervention protocol appears safe and feasible for adolescents in the postacute stage of sports-related concussion recovery.

A further meta-analysis was performed of the 5 studies that reported postintervention PCSS results. Interestingly, the data were homogeneous. This was quite rational as all 5 studies employed a similar intervention (active intervention or occupational therapy). Fig. 3 reveals that active rehabilitation improved the PCSS score by 4.14 points, a statistically significant difference (P≤0.05). The I² homogeneity test revealed homogeneity; therefore, a fixed-effects model was used. Fig. 4 shows that no publication bias existed in atleast 5 studies describing chronic intervention using rehabilitation or occupational therapy.

The risk of bias assessment is shown in Fig. 5. Of the 16 RCTs assessed using the RoB 2, 10 had a low risk of bias, 4 had some concern of bias, and 2 had a high risk of bias in at least 1 domain. The results are summarized in Fig. 5.

Based on the findings of the current and previous RCTs, rest is important in the acute management of pediatric patients [20,24]. Two RCTs recommended strict rest for 5 days as an intervention in pediatric patients who had symptoms following acute concussion. However, strict rest did not improve outcomes and may have contributed to increased daily symptoms and slower symptom resolution. Rather, the best strategy is the current usual care (rest for 1–2 days with a stepwise return to activity). However, the International Concussion in Sports Group and the Centers for Disease Control and Prevention recommend complete cognitive and physical rest for the first 24–48 hours after sustaining a concussion, followed by a structured return to activity. The manner in which activities are considered cognitive rest during the acute recovery period is subject to variations in practice and guidelines, e.g., screen time abstinence is often interpreted as an example of cognitive rest.

Concussions cause a cascade of complex, overlapping, and disintegrative processes in the brain. These changes are caused by numerous pathophysiological processes, inluding ionic fluctuations, neural structural damage, neuroinflammation, an increased release of extracellular neurotransmitters, and reduced cerebral blood flow. Although recent human studies have shown that these processes may return to normal, the recovery of each may occur over a specific period and is probably affected by several factors [30].

Adequate recovery time following concussion should occur; however, early integration into regular physical activity that does not worsen outcomes should be considered to facilitate a proper healing environment. The possible adverse effects of prolonged activity restrictions should be considered when prescribing rest for more than a few days, such as in school-aged children. It is logical that a long rest period, especially in adolescents, can lead to deconditioning, which in turn can lead to changes in cerebrovascular control. However, these changes lead to other symptoms independent of the initial injury [31].

Thirteen RCTs investigated the effects of nonpharmacological treatments on concussions. Studies have reported the efficacy of aerobic exercise, active rehabilitation, vestibular rehabilitation, virtual reality, family intervention, hyperbaric oxygen therapy, collaborative care, soft-tissue release, mobilization and/or thrust manipulations, oculomotor control, neuromotor control, and balance exercises.

Based on these findings, aerobic exercise is the most commonly reported nonpharmacological approach to managing concussion. Three reports showed that patients assigned to aerobic exercise (intervention group) had significant symptom improvement compared with those assigned to stretching (control group) [22,25,27]. All 3 studies had a low risk of bias; nonetheless, 2 were relatively small. Moreover, all studies used different methods of aerobic exercise, interventions, and follow-up durations.

Two RCTs used active rehabilitation consisting of aerobic training, coordination exercises, visualization, and imagery techniques with a physiotherapist [16,28]. Chan et al. [16] showed the safety, tolerability, and potential efficacy of active rehabilitation for adolescents with persistent postconcussive symptoms. According to Dobneyetal. [28], active rehabilitation can be started as soon as 2-week postinjury.

The efficacy of vestibular rehabilitation was demonstrated in one RCT in the context of vestibular dysfunction [19]. Vestibular rehabilitation uses central mechanisms of neuroplasticity (adaptation, habituation, sensorial, and functional substitution) to increase static and dynamic postural stability and improve visual–vestibular interactions [32].

Four RCTs found that visual rehabilitation improved visual symptoms. Vision therapy improves communication between the eyes and brain through various exercises led by the provider. Vision therapy involves a range of techniques aimed at promoting binocular vision and enhancing visual processing, tracking, and perception using diverse approaches and exercises. Most of these therapies lack specificity, and a subset of vision therapy practitioners assert that these visual impairments serve as underlying factors for numerous learning problems observed in youngsters [23]. The implementation of vision therapy can enhance several visual skills, including eye tracking, focus, eye-teaming, depth perception, and hand-eye coordination.

The strength of our study was its limited meta-analytical design. We demonstrated the interesting roles of rehabilitation and occupational therapy in pediatric concussion management. We also identified several interesting nonpharmacological approaches. This was an improvement on the scoping review performed by Dobney et al. in 2017 [33]. We confirmed the conclusion of Dobney in that anaerobic exercise is a promising concussion treatment strategy. Finally, further research with robust methodologies, larger samples, and a multicenter design is necessary to enhance the level of evidence.

The main limitation of this review was the notable interstudy heterogeneity in the outcome measures, diagnosis, duration of intervention, and treatment, which limited the meta-analysis. Nevertheless, we were able to pinpoint a possible approach to performing a meta-analysis that revealed an interesting finding. In addition, most of the studies were performed at single centers. The diagnosis of concussion is largely based on clinical competence, which includes an assessment of the patient's history of present illness, a comprehensive systems review, and a thorough physical examination. Numerous assessment tools were used in these studies, such as the Sports Concussion Assessment Tool (SCAT), now in its fifth edition, and the associated Child SCAT5 (for use for children aged 5–12 years). These tools may be used on the sidelines or in the clinic to support or refute the diagnosis of concussion. Although such essential tools may greatly assist decisions, they are neither essential nor conclusive for diagnosing concussions. Moreover, the eligibility criteria were limited to studies published in English and a comprehensive search being conducted of only 4 databases.