Neuro-Ophthalmologic Response to Repetitive Subconcussive Head Impacts: A Randomized Clinical Trial

IMPORTANCE: Subconcussive head impacts have emerged as a complex public health concern. The oculomotor system is sensitive to brain trauma; however, neuro-ophthalmologic response to subconcussive head impacts remains unclear. OBJECTIVE: To examine whether subconcussive head impacts cause impairments in neuro-ophthalmologic function as measured by the King-Devick test (KDT) and oculomotor function as measured by the near point of convergence. DESIGN, SETTING, AND PARTICIPANTS: In this randomized clinical trial, adult soccer players were randomized into either a heading group or kicking (control) group. The heading group executed 10 headers with soccer balls projected at a speed of 25 mph. The kicking-control group followed the same protocol but with 10 kicks. Peak linear and rotational head accelerations were assessed with a triaxial accelerometer. The KDT speed and error and near point of convergence were assessed at baseline (preheading or prekicking) and at 0, 2, and 24 hours after heading or kicking. EXPOSURES: Ten soccer-ball headings or kicks. MAIN OUTCOMES AND MEASURES: The primary outcome was the group-by-time interaction of KDT speed at 0 hours after heading or kicking. The secondary outcomes included KDT speed at 2 hours and 24 hours after heading or kicking, KDT error, and near point of convergence. RESULTS: A total of 78 individuals enrolled (heading group, n = 40; kicking-control group, n = 38). Eleven individuals (heading group: 4 women; mean [SD] age, 22.5 [1.0] years; kicking-control group, 3 women and 4 men; mean [SD] age, 20.9 [1.1] years) voluntarily withdrew from the study. Data from 67 participants with a mean (SD) age of 20.6 (1.7) years were eligible for analysis (heading, n = 36; kicking-control, n = 31). Mean (SD) peak linear accelerations and peak rotational accelerations per impact for the heading group were 33.2 (6.8) g and 3.6 (1.4) krad/s2, respectively. Conversely, soccer kicking did not induce a detectable level of head acceleration. Both groups showed improvements in KDT speed (heading group: 0 hours, −1.2 [95% CI, −2.2 to −0.1] seconds; P = .03; 2 hours, −1.3 [95% CI, −2.6 to 0] seconds; P = .05; 24 hours, −3.2 [95% CI, −4.3 to −2.2] seconds; P