In Stride with Sports Medicine

Research suggests concussed athletes may return too soon

Published in the April 2008 issue of BioMechanics

by Jordana Bieze Foster


Current discussions of return-to-play criteria for athletes recovering from concussion tend to revolve around neuropsychological testing. But research from the University of Oregon suggests that such criteria should also take into account dynamic motor tasks, which may remain impaired even after neuropsychological measures have returned to pre-injury levels.

“Motor stability, particularly balance control when attention is divided, can be impaired for up to one month following what is sometimes considered to be a mild concussion,” said Li-Shan Chou, PhD, associate professor of human physiology at the university and senior author of the study, which was published January 31 in the online version of Medical Engineering and Physics. “Hence, neuropsychological assessments alone may not be adequate to detect lingering insufficiencies in motor function that could influence sports performance and vulnerability to further injury.”

Chou and colleagues also found that athletes who participate in contact sports were more likely to demonstrate impaired gait and balance than non-athletes, even in cases where the athletes had not been diagnosed with concussion—suggesting that head trauma may be affecting athletes' motor skills long before symptoms of concussion become evident.

“It is possible that diagnosed concussions as well as repeated exposure to sub-concussive blows in the non-concussed and concussed athlete groups decreased the ability of these individuals to control their sway during gait,” Chou said.

The researchers analyzed 28 concussed individuals and 28 matched controls; half of the subjects in each group were NCAA Division 1 or university club level athletes who played football, rugby, or lacrosse. The gait stability protocol required the subjects to walk along a 10 meter walkway at a self-selected speed, while investigators measured whole-body center of mass displacement and velocity as well as gait velocity, under single- and dual-task conditions. Testing was done 48 hours after injury, and again at five, 14 and 28 days post-concussion.

The investigators found clearly different gait stability patterns, in both the frontal and sagittal planes, that differentiated athletes from non-athletes, regardless of concussion status. Athletes demonstrated significantly greater sway excursion, faster sway velocity, slower gait velocity, and a smaller amount of separation between center of mass and center of pressure separation before each step. These differences were more pronounced under the dual-task condition than the single task condition, and persisted for the 28-day duration of the study.

In an earlier study, published in the December issue of the British Journal of Sports Medicine, the same researchers found that neuropsychological measures initially differentiated 29 concussed subjects from non-injured matched controls, but had resolved in most cases by day 14 and in all cases by day 28 after injury. By contrast, medial-lateral displacement during the dual-task condition remained significantly higher for the concussion group than the non-injured group at 28 days, although there was no significant between-group difference for the single-task condition.

The University of Oregon findings are particularly interesting in light of previous research suggesting that postural stability, as measured using the Sensory Organization Test in conjunction with NeuroCom balance testing devices, tends to resolve within three or four days following concussion—much sooner than neuropsychological measures. Compounding the issue is the fact that, because of the multifactorial nature of concussion and its manifestations, not all concussed athletes demonstrate postural stability impairments.

Investigators from the University of North Carolina at Chapel Hill, however, reported in a September 2006 Journal of Athletic Training study that nonlinear dynamics could be used to identify persistent postural control deficits in concussed athletes—even those who did not appear to be unstable.

The researchers analyzed 29 athletes who had sustained concussions while playing football, soccer, lacrosse or field hockey. Sixteen subjects, whose equilibrium scores (a composite of SOT test results) at 48 hours after injury were no more than 5% below their preseason scores, were classified as “steady”; 13 were classified as “unsteady.”

As expected, equilibrium scores returned to near preseason levels within 48 to 96 hours post-injury in the unsteady group. However, approximate entropy (a nonlinear dynamics estimate of the randomness of COP data) measures indicated that COP displacement in the medial-lateral direction was significantly less random following injury in all subjects, steady or unsteady, and that this significant difference was maintained at 96 hours post-injury—even after equilibrium scores had resolved.

“Approximate entropy is detecting something we can't see very well with our eyes. Just because these athletes look steady, doesn't mean all is entirely normal,” said Jim T. Cavanaugh, PhD, first author of the study, who is now an assistant professor of physical therapy at the University of New England in Portland, ME. “It's one more reason to be conservative with the decision to return these athletes to the field.”


Copyright 2008 Jordana Foster – 24 Kirkland Dr, Stow, MA – Email: – Fax: (815) 346-5239