Accommodative Amplitude and Pupillary Light Reflex Metrics Before and After Exercise Among Adolescents with Sport-related Concussion

Background: Exercise intolerance is a manifestation of autonomic nervous system (ANS) dysfunction after concussion. Accommodative amplitude (AA) and pupillary light reflex (PLR) may be useful clinical proxies for ANS dysfunction.

Methods: Secondary analysis of data from a randomized controlled trial of aerobic exercise after adolescent sport-related concussion (SRC). Monocular accommodation was measured before and after the Buffalo Concussion Treadmill Test (BCTT). Fisher’s exact tests were used to compare AA before and after exercise, while Wilcoxon rank sum was used to assess the relationship between AA and time until clinical recovery. In a subset of 11 participants, pupillary light reflex metrics were obtained before and after exercise using a hand-held, infrared pupillometer device.

Results: 99 participants (64.6% male, median age 15.7 (IQR 14.5, 16.8)) had AA measured before and after the BCTT at the initial post-injury visit (median 6 days (IQR 4, 8)). Abnormal AA, defined using Hofstetter’s age-adjusted formula, was present in 43 (43.4%) pre-exercise and in 45 (45.5%) post-exercise. Abnormal pre- and post-exercise AA were significantly associated (p < 0.001). AA improved with exercise for 9 participants (i.e., changed from abnormal to normal) while it worsened with exercise for 11 (i.e., normal to abnormal). Low heart rate threshold (HRt), defined as HR less than 135 bpm at test cessation due to symptom exacerbation, was significantly associated with abnormal post- (p = 0.005) but not pre-exercise AA (p = 0.225). Participants with abnormal post-exercise AA had significantly longer time to clinical recovery (median 20.5 days (IQR 14.5, 32)) than those with normal post-exercise AA (median 14 days (IQR 10.75, 21.75)) (p = 0.008).

Among those participants that completed pupillometry testing at initial visit, percent constriction and maximum constriction velocity parameters were significantly reduced after exercise compared to before (p < 0.05 after Holm’s correction). A similar trend was seen at all follow-up visits; however, pre- and post-exercise values were not significantly different, suggesting at least partial recovery of ANS function within the first 1-2 weeks of injury.

Conclusion: Adolescents with reduced exercise tolerance after concussion are more likely to have abnormal AA after exercise, suggesting that concussion-related autonomic dysfunction is precipitated by an aerobic exercise stress test. Abnormal AA and PLR metrics after exercise may provide prognostic utility in concussion. Since monocular accommodation is essential for successful participation in school, detecting deficits in AA that are due to ANS dysfunction and require targeted exercise rehabilitation rather than vision therapy may help with the return to school process after concussion.