Single-Legged Hop and Single-Legged Squat Balance Performance in Recreational Athletes With a History of Concussion

A convenience sample of 30 college-aged recreational athletes (no varsity intercollegiate athletes were included) was recruited and separated into 2 groups of 15 participants: (1) concussion group (median time since concussion = 126 days [range = 28–432 days]) and (2) control group with no recent concussions matched to each injured participant on the basis of sex, age (±1 year), height (±5%), and mass (±10%). All volunteers self-reported completing moderate to intense physical activity for >30 consecutive minutes ≥3 times per week. Volunteers were excluded if they self-reported attention-deficit/hyperactivity disorder, seizure disorder, lower extremity injury resulting in ≥3 days missed from physical activity within the 6 months before the study, any history of lower extremity or low back surgery, concussion requiring hospital admittance, any concussion symptoms at the time of the study, or a history of >3 concussions. Participants in the concussion group must have sustained a concussion diagnosed by a medical professional within the 1.5 years before the study, whereas participants in the matched control group must have been without a diagnosed concussion for ≥3 years. All participants provided written informed consent, and the study was approved by The University of North Carolina at Chapel Hill Institutional Review Board.

Kinetic data and the center of pressure (COP) were sampled at 1200 Hz from 2 force plates (Bertec Corp, Columbus, OH) embedded in the laboratory floor. For the single-legged hop, participants were outfitted with a single retroreflective marker over the superior sacrum. A Vicon Motion Capture System (Centennial, CO) recording at 120 Hz tracked the sacral marker position during the single-legged squat for calculation of squat displacement (depth) and speed.

Participants stood atop a 30-cm box placed at a horizontal distance equal to 50% of their height behind a force plate. With both hands on their hips, they jumped off the box with both feet, and landed on a single limb. Participants were instructed to achieve a stable position as quickly as possible upon landing. Trials were discarded and repeated if the participant touched down with the opposite foot or if any portion of the stance foot left the force plate after ground contact. The total trial time was 10 seconds, with participants holding their best single-legged posture for as much of the trial as possible. They performed 5 single-legged hop trials on each limb in random order.

Participants were instructed to stand on a single limb with their toes facing forward. They were positioned with the non–weight-bearing limb flexed to 90° at the knee and approximately 75° at the hip, their hands on their hips, and their head and eyes facing forward. Participants flexed their weight-bearing knee into a squat position to maximum comfort and returned to the upright posture. They performed 5 single-legged squat trials on each limb. Each trial consisted of 5 sequential squat repetitions completed without pause at a self-selected pace.

A fourth-order, low-pass Butterworth filter with a cutoff frequency of 10 Hz was applied to all kinematic data. For the single-legged hop, kinetic data were filtered using a second-order, low-pass Butterworth filter with a cutoff frequency of 12.53 Hz.²⁰  For single-legged squat COP, kinetic data were filtered using a fourth-order, low-pass Butterworth filter with a cutoff frequency of 14 Hz.¹⁹  Several commonly reported balance measures were calculated during the single-legged squat and are defined in Table 1.

Time to stabilization (TTS) was calculated as the total time participants took to establish stable balance on their landing limb after the single-legged hop. The methods for measuring TTS have been described in detail elsewhere.¹⁸  Using a custom MATLAB script (The MathWorks, Natick, MA), we calculated normalized (to body weight) resultant ground reaction force during the last second of each trial, when participants were believed to have established stable balance on their landing limb, by squaring each time-series value of the anterior-posterior and medial-lateral ground reaction forces, summing the values, and taking the square root of the sum.^21,22  The group mean and standard deviation of the control group only were calculated to determine the resultant ground reaction force range of variation. Three standard deviations were added to the mean range of variation, and this value was multiplied by each participant's body weight to calculate a normalized reference value for each individual in both groups. This normalized reference value represented the standardized resultant vertical ground reaction force associated with stable balance for each participant. Using a custom LabVIEW program (National Instruments, Austin, TX), we fit the entire resultant ground reaction force from the trial with an unbounded third-order polynomial. Time to stabilization was defined as the time when the unbounded third-order–polynomial resultant ground reaction force fit curve fell below the normalized reference value.

Data were averaged across all trials for a given limb, and these average values were used for all statistical analyses. The dominant leg was defined as the self-reported limb used to kick a soccer ball for maximum distance. Dependent variables were calculated separately for the dominant and nondominant legs and were analyzed in SAS (version 9.4; SAS Institute, Cary, NC) using between-participants analysis of covariance. We covaried for the number of days between the last concussion and the testing session in all statistical models. The number of days postinjury for each concussion-group participant was subtracted from the group mean days since concussion (177 days). Control-group participants were assigned a value of zero.¹⁴  This created a value representing mean-centered days since concussion, which was used as a covariate in all statistical models. An a priori α level of .05 was established.

We observed no between-groups differences in participant characteristics. Median days since concussion in the concussion group was 126 (range = 28–432). One participant in the control group reported a history of 2 concussions, the most recent of which was 1103 days (>3 years) before testing. No other control-group participants reported a history of concussion. Group characteristics and between-groups comparisons are shown in Table 2.

Time to stabilization on the dominant leg did not differ between the concussion (2.04 seconds; 95% confidence interval [CI] = 1.65, 2.43 seconds) and control (1.84 seconds; 95% CI = 1.49, 2.19 seconds) groups (F_2,27 = 0.64, P = .43), but the concussed group (1.90 seconds; 95% CI = 1.68, 2.12 seconds) took longer to stabilize than the control group (1.55 seconds; 95% CI = 1.33, 1.77 seconds) during nondominant leg TTS (mean difference = 0.35 seconds; 95% CI = 0.04, 0.64 seconds; F_2,27 = 5.69, P = .02). The TTS values are presented in the Figure.

No differences were observed for any of the outcomes measured during the single-legged squat for the dominant (P ≥ .11) or nondominant (P ≥ .21) leg. All descriptive statistics and statistical comparisons for the single-legged squat are provided in Table 3.

Numerous researchers^23–25  have suggested that dynamic-balance deficits during gait may persist beyond the return to play after concussion. These deficits may contribute to an increased musculoskeletal injury risk after concussion,^9–12  but more study of the dynamic aspects of movement is needed. Our findings indicated that TTS after a single-legged hop may distinguish individuals with a history of concussion from healthy control participants. However, we observed no between-groups balance differences for the single-legged squat.

Time-to-stabilization deficits have been identified in individuals with chronic ankle instability^18,26  or recent anterior cruciate ligament injury²⁷  and in female athletes at an average of 2.5 years after anterior cruciate ligament reconstruction.²²  Although seemingly small, our observed difference of approximately 0.35 seconds was similar to previously reported^22,26  group differences between individuals with and those without chronic ankle instability or anterior cruciate ligament reconstruction. Furthermore, TTS measures can be used to accurately discriminate between individuals with and those without chronic ankle instability.²⁶  Therefore, TTS has clinical utility in identifying individuals who may be at risk for ankle injuries before or during athletic participation. Targeted injury-risk mitigation programs can be designed to strengthen the ankle joint and increase proprioception with the intention of limiting acute and chronic ankle injuries. Further research into TTS after concussion is important to determine whether it may have similar clinical applicability.

We did not find any group differences in COP characteristics during the single-legged squat. This task may not have challenged the postural-control system sufficiently to demonstrate between-groups differences in our study because the injured group was tested an average of 177 days after concussion. Future investigators should consider using acute postconcussion testing and adding a cognitive component to the single-legged squat, given that dynamic-balance deficits after concussion appeared to be more pronounced under dual-task conditions.²⁸  Dynamic dual-task assessments in a research setting may provide an increased challenge for the postural-control system that better reflects true sport demands.¹⁹ 

Although concussion does not damage the peripheral somatosensory organs or muscle tissue as an ankle or knee sprain does, it may affect central processing of peripheral somatosensory information. Researchers have suggested that cortical excitability decreases²⁹  and motor-evoked potential latency increases after concussion.³⁰  If somatosensory information is not processed quickly or accurately by the brain, motor output could be slowed or perturbed, leading to aberrant movement patterns during gait and dynamic movement in sport. Powers et al²⁹  proposed that concussed individuals may recruit fewer muscle fibers during a maximum contraction than matched control participants. Voluntary muscle activation is strongly related to strength.³¹  Therefore, concussed individuals may have a reduced ability to efficiently and quickly process somatosensory information and may struggle to appropriately respond to external stimuli with a powerful muscle contraction. In a highly dynamic environment, such as sport, even minor central-processing deficiencies or limited muscle recruitment could increase the musculoskeletal injury risk.

Whereas static balance is commonly assessed after a suspected concussion and throughout recovery, dynamic-balance assessments are not as common. Time to stabilization may allow for a more comprehensive assessment of the postural-control system in environments that more closely mimic those the athlete will experience during sport participation. Based on our observed TTS deficits, we suggest that lingering neuromuscular-control deficits may persist beyond the time when athletes traditionally are cleared to return to play. We are the first to explore TTS in individuals with a history of concussion and believe our findings underscore the need for further longitudinal studies in this area, especially because we observed no between-groups differences during the single-legged squat. An understanding of how dynamic-balance outcomes, such as TTS, are affected acutely after concussion may benefit clinical practice. Understanding the recovery of dynamic balance may result in more comprehensive and safer return-to-play testing and criteria, potentially negating undesirable consequences, such as an increased musculoskeletal injury risk.

Our study had several limitations. We observed wide variability in the time between concussion and the testing session, which could have skewed our results. We controlled for this variability in our statistical analyses, but our relatively small sample size may have minimized our ability to find larger between-groups differences. Some of our measures may have been affected by preexisting conditions, such as chronic ankle instability.^18,26  Whereas our criteria were designed to exclude participants with musculoskeletal, surgical, or central nervous system conditions that could affect outcomes, participants may have had conditions that we were unaware of at the time of testing. Our findings in this sample should be confirmed in more prospective, larger-scale investigations. Also, we studied recreational athletes. Their involvement in specific sports varied, and in many cases, our participants reported playing multiple sports. Differences in sports training, experience, and collision-sports history could have influenced the dependent variables we measured. Inferring any such influences would be speculative given the lack of empirical data supporting such statements. Sport-specific cohorts should be investigated, and tasks that are specific to movements occurring more frequently in individual sports should be assessed. Examples include anticipated and unanticipated cutting, jumping for maximal vertical height or horizontal distance, and reacting to visual or auditory stimuli while cutting or jumping.

Individuals with a history of concussion took longer to stabilize during a single-legged hop task on their nondominant legs. Given the preliminary nature of our work, the clinical utility of our findings is unclear. Researchers should develop methods that allow them to assess these variables acutely after injury and identify how dynamic balance recovers after concussion. Understanding the recovery of dynamic balance in relation to more traditional concussion-assessment measures may greatly enhance the care afforded to individuals who sustain a brain injury. Beyond dynamic-balance assessment, examining other outcomes, such as joint kinetics, stiffness, reaction time, and muscle activation, may provide important insights into neuromuscular deficits postconcussion.

Static balance is commonly assessed after concussion, and deficits have been described using static assessments. However, during sport participation, the neuromuscular system is challenged further, as evidenced by postconcussion balance-control deficits in gait that linger well beyond the recovery of static balance. In addition, the musculoskeletal injury risk is increased after concussion, but the underlying mechanisms for this increased risk are unknown. The observed nondominant-leg TTS deficits in a recently concussed cohort as compared with matched control individuals may provide evidence for neuromuscular deficits that are not currently addressed after concussion. Clinicians should consider incorporating more dynamic-balance assessments after concussion, although additional research is needed to determine their prognostic value.

This study was funded by The University of North Carolina at Chapel Hill Junior Faculty Development Award (Dr Mihalik). The views expressed in this paper are our own and do not reflect the official policy of the Departments of the Army, Navy, and Air Force; Department of Defense; the Uniformed Services University of the Health Sciences; or the US government.