The Effect of Physical Exercise After a Concussion: A Systematic Review and Meta-analysis

Avtar Lal,Stephanie A. Kolakowsky-Hayner,Jamshid Ghajar,Maya Balamane

DOI: https://doi.org/10.1177/0363546517706137

IF: 4.8

2017-06-01

The American Journal of Sports Medicine

Abstract:Background: Data evaluating the role of exercise in patients with a concussion are contradictory. Studies have reported improvement in the Post-Concussion Symptom Scale (PCSS) score, whereas others showed no effect on the PCSS score. Purpose: To conduct a systematic review and meta-analysis on the role of physical exercise on different outcomes in patients with a concussion. Study Design: Systematic review and meta-analysis. Methods: A search of 5 databases from the earliest available date to September 30, 2016, and a hand search of a few articles were performed. Trial registries were reviewed, and authors of multiple studies were contacted to find additional published or unpublished studies. Randomized controlled trials (RCTs), cohort studies, and before and after (pre-post) studies evaluating the effect of physical exercise, compared with control, in patients with a concussion or mild traumatic brain injury were included. Results: The search generated 1096 studies. Of these, 14 studies (5 RCTs, 1 propensity score matching study, 3 cohort studies, and 5 before and after studies) met our inclusion criteria. Exercise significantly decreased the PCSS score (mean difference, −13.06; 95% CI, −16.57 to −9.55; P < .00001; I 2 = 44%), percentage of patients with symptoms of a concussion (risk ratio, 0.74; 95% CI, 0.63 to 0.86; P = .0001; I 2 = 0%), and days off work (17.7 days vs 32.2 days, respectively; P < .05) compared with control. Exercise improved the reaction time (standard mean difference, –0.43; 95% CI, −0.80 to −0.06; P = .02) component of the Immediate Post-Concussion Assessment and Cognitive Testing (ImPACT) score without affecting the Balance Error Scoring System (BESS) score and neuropsychological parameters. The Grading of Recommendations Assessment, Development and Evaluation (GRADE) scores were moderate for the PCSS, symptoms, ImPACT, BESS, and neuropsychological tests. Conclusion: Physical exercise appears to improve the PCSS score and symptoms in patients with a concussion. A high-quality RCT evaluating different intensities of exercise at different time points, for different durations after a concussion, for different races/ethnicities, and for sex needs to be conducted to evaluate a clear effect of exercise in patients with a concussion.