Clinical Scenario

Historically, adolescent athletes with sport-related concussion (SRC) have been advised to abstain from physical activity until asymptomatic, at which point they may begin a graduated return to sport protocol.¹ Some researchers, however, have proposed that subsymptom exacerbation low-intensity aerobic exercise (LIAEX) may expedite concussion recovery via increased levels of brain-derived neurotrophic factor.^2–5 Subsymptom exacerbation LIAEX, starting a minimum of 4 weeks after an SRC, has been reported to be more beneficial than rest,⁶ but the effects of LIAEX earlier than 4 weeks need to be elucidated. This review was conducted to determine if evidence supports initiating subsymptom exacerbation LIAEX within the first 4 weeks following SRC. The review was limited to adolescent athletes (aged 10–19 y), because, historically, this group experiences high rates of SRC and appears to be more vulnerable to prolonged recovery.^1,7

Focused Clinical Question

Does subsymptom exacerbation LIAEX within the first 4 weeks following SRC hasten symptom resolution in adolescent athletes?

Summary of Search, “Best Evidence” Appraised, and Key Findings

1. •The literature was searched for studies on the effects of early (within 4 wk) subsymptom exacerbation LIAEX after medically diagnosed SRC in adolescents.
2. •The search yielded 3 articles^2,7,8 that fit inclusion/exclusion criteria. One level 1b randomized controlled trial (RCT),² and 1 level 2b nonrandomized trial (NRT)⁷ compared LIAEX initiated within 10 days after SRC with usual care (ie, rest). A second level 1b RCT⁸ compared LIAEX initiated within 10 days after SRC with placebo stretching.
3. •Recovery in the studies was defined as symptom resolution per the Postconcussion Symptom Scale (PCSS) and/or examination by a physician.
4. •One RCT⁸ found a significant difference in days to recovery favoring the LIAEX group compared with the stretching group. Likewise, the NRT by the same group of investigators showed faster recovery with LIAEX versus usual care (ie, rest). By contrast, the other RCT² did not find a significant between-group difference in days to recovery, although this study appears to have been underpowered.
5. •Notably, there were no reports of adverse effects with LIAEX in any of the reviewed studies.

Clinical Bottom Line

Regarding the posed clinical question, evidence from 1 level 1b RCT⁸ and 1 level 2b NRT⁷ suggests subsymptom exacerbation LIAEX started within 10 days after SRC is associated with a faster overall recovery time when compared with rest or placebo stretching, although additional data are needed to substantiate this conclusion. Moderately strong (level 1 and 2) evidence from all 3 studies indicates that SRC symptom severity decreased across the study interval in response to LIAEX. In the reviewed studies, no adverse effects were reported when LIAEX was performed 5 to 7 days per week on a stationary bike or treadmill for 10 to 20 minutes at 50% to 70% of age-predicted maximal heart rate² or at 80% of symptom-threshold heart rate.^7,8

Strength of Recommendation

Level 1 evidence from 2 RCTs and level 2 evidence from 1 NRT support early (<10 d) subsymptom exacerbation LIAEX for adolescents with SRC. It is important to acknowledge that the RCT by Micay et al² in this review appears to have been underpowered with a total sample size of 15. This small sample may have prevented detection of between-group differences. The NRT by Leddy et al^8,9 compared convenience sample data from 2 different published studies, and thus group assignment was not random. Finally, the RCT by Leddy et al⁸ did not a have a true control group, rather, a placebo stretching group that was compared with the LIAEX group. Given the limited number of studies on this topic and the methodological shortcomings described earlier, additional data from studies with greater methodological quality and homogeneity are advised before widespread adoption should be advocated.

Search Strategy

Inclusion and Exclusion Criteria

Results of Search

The initial search in October 2018 yielded 24 studies that were reviewed for appropriateness based on inclusion and exclusion criteria. Of these, 2 articles^2,7 were selected for critical appraisal. An additional article, published in February 2019, was identified on PubMed.⁸ A summary of the selected articles is provided in Tables 1 and 2.

Best Evidence

The studies in Tables 1 and 2 were identified as the best available evidence for this review. Two studies were level 1b evidence RCT and 1 study was a level 2b evidence NRT. Levels of evidence for this review were based on the Centre for Evidence-Based Medicine 2009 criteria.¹⁰

Implications for Practice, Education, and Future Research

Using LIAEX after SRC is a relatively recent practice, and, accordingly, the safety and efficacy of this approach need to be well established by high-quality research before widespread clinical adoption should be considered. The reviewed studies reported no adverse effects in symptomatic adolescents following subsymptom exacerbation LIAEX started within 10 days post-SRC. In the NRT, none of the 24 participants in the LIAEX group had delayed recovery, while 4 of the 30 (13%) rest group participants had delayed recovery.⁷ In addition, in the RCT by Leddy et al,⁸ 2 of the 52 (3.8%) participants with acute SRC in the LIAEX group had delayed recovery, while 7 of the 51 (13.7%) stretching group participants had delayed recovery. Therefore, a potentially important advantage of exercise-based treatment early in the SRC recovery process may be the reduced incidence of delayed recovery.

Regarding the efficacy of subsymptom exacerbation LIAEX for concussed adolescents, all 3 studies reported a reduction in SRC symptom severity across their study intervals. The magnitude of symptom reduction across the recovery timeline was greater in the LIAEX group than the rest group in the NRT,⁷ but not the 2 RCTs.^2,8 Reviewers calculated effect sizes for between-group symptom severity reduction (via PCSS scores) from the data provided in the Micay et al² RCT; effect sizes were considered clinically significant (2.8 between weeks 1 and 3; 1.6 between weeks 1 and 4), favoring the LIAEX group. In the NRT, adolescents who started LIAEX within 9 days of SRC had greater symptom reduction across the study period than the usual care (ie, rest) group. They also reported the LIAEX group had a significantly shorter recovery time than the usual care group, although their control data set was imported from another study, and group assignment was not randomized.⁷ By contrast, the RCT by Micay et al did not detect between-group differences in recovery time, although their small sample size (N = 15) may have lacked the statistical power to detect differences. According to Micay et al,² a calculated sample size of 30 would be needed to detect between-group differences with an effect size of 0.8. Calculated effect sizes for reduction in time to medical clearance were moderate in both the Micay et al² and Leddy et al⁷ studies (0.4 and 0.5, respectively). In the RCT by Leddy et al,⁸ both groups reported decreases in PCSS scores through the 21 days of the study; however, when compared with the stretching group, the LIAEX group reported significant PCSS score decreases on days 5, 12, and 18. Additional RCTs are needed to definitively answer the clinical question regarding the impact of subsymptom exacerbation LIAEX on overall SRC recovery time in this population, although the initial data are encouraging.

Clinicians should educate patients that aerobic exercise is beneficial for the human brain and that subsymptom exacerbation LIAEX does not appear to delay recovery, but it may speed recovery in adolescents with acute postconcussive symptoms. This seems concordant with a recent international consensus statement that recommends a more active approach to SRC treatment.¹ There is evidence that aerobic exercise can promote neuron growth and repair after concussions through increased levels of brain-derived neurotrophic factor.⁵ Levels of brain-derived neurotrophic factor have been shown to increase in healthy athletes immediately after cycle ergometry to exhaustion¹¹ and in rodents that initiated exercise within 1 to 3 days after concussion.¹² In the later study, concussed rodents that were exercised also had significant improvements in motor and cognitive functioning. Current evidence points to inferior neurocognitive performance of athletes engaging in high levels of activity following SRC when compared with athletes engaging in moderate levels of activity.¹³ Given this, clinicians should emphasize to their patients that any exercise or activity should be below the threshold of symptom exacerbation.

Future research on this topic would be strengthened by: (1) using mixed-gender participant groups, (2) including a control group and random group assignment for interventional comparisons, (3) determining optimal sample sizes a priori via a power analysis, (4) employing a standardized LIAEX protocol, (5) using standardized outcomes measures to determine readiness to return to sport/activity (ie, medical clearance), and (6) reporting effect sizes so the clinical significance can be appreciated. The RCT by Micay et al² and the NRT by Leddy et al⁷ included male subjects only for convenience.’ Females, however, may be more likely to have a delayed recovery after SRC.¹⁴ As with the reviewed RCT by Leddy et al,⁸ inclusion of both males and females in future studies will improve generalizability. The consistent utilization of randomly assigned control or comparator groups will aid in estimating the influence of aerobic exercise post-SRC. In addition, sample size calculations should be performed to ensure studies on this topic are adequately powered.

There is a need to determine optimal exercise protocols that could be adopted by future researchers. Two of the studies in this review utilized daily subthreshold aerobic exercise lasting 20 to 30 minutes,^7,8 while the third study used a prescribed progressive aerobic exercise program, based on age-predicted maximal heart rate, beginning at 10 minutes and increasing over time to 20 minutes.² In both articles by Leddy et al,^7,8 the heart rate at symptom exacerbation was used to determine a target subsymptom threshold heart rate for aerobic exercise. Determination of optimal LIAEX dosage is needed, as 2 studies implemented daily exercise^7,8 and the other prescribed exercise 5 days per week.² Ongoing research to develop a standardized aerobic exercise protocol for concussed patients is being conducted.¹⁵ Future researchers should aim to standardize outcome measures; all 3 of the reviewed studies included the PCSS as an outcome measure, although this is a highly subjective, symptom-based test. Furthermore, in the studies by Leddy et al,^7,8 symptom recovery was defined as a PCSS score of <7 for 3 consecutive days, while Micay et al² relied solely on physician examination to determine recovery.² Thus, there is a need for more standardization in determining recovery and readiness for return to sport after SRC. As with many health care issues, greater methodological consistency in future studies should improve confidence in the validity of results and enhance clinical decision making.

The results of this review suggest subsymptom exacerbation LIAEX administered within 10 days following SRC in adolescents does not appear to delay recovery and may result in greater symptom reduction and faster medical clearance than rest. Clinicians who consider applying the LIAEX protocols, as described in the clinical bottom line, should emphasize to their patients that any exercise or activity should be below the threshold of symptom exacerbation. Information in this critically appraised topic should be reviewed in 3 years or after the publication of high-quality RCTs on this topic to determine if the recommendations suggested earlier should be modified.