Abstract
US Soccer eliminated soccer heading for youth players ages 10 years and younger and limited soccer heading for children ages 11–13 years. Limited empirical evidence associates soccer heading during early adolescence with medium-to-long-term behavioral deficits. The purpose of this study was to compare sensory reweighting for upright stance between college-aged soccer players who began soccer heading ages 10 years and younger (AFE≤10) and those who began soccer heading after age 10 (AFE>10). Thirty soccer players self-reported age of first exposure (AFE) to soccer heading. Sensory reweighting was compared between AFE≤10 and AFE>10. To evaluate sensory reweighting, we simultaneously perturbed upright stance with visual, vestibular, and proprioceptive stimulation. The visual stimulus was presented at two different amplitudes to measure the change in gain to vision, an intra-modal effect; and change in gain to galvanic vestibular stimulus (GVS) and vibration, both inter-modal effects. There were no differences in gain to vision (p=0.857, η2=0.001), GVS (p=0.971, η2=0.000), or vibration (p=0.974, η2=0.000) between groups. There were no differences in sensory reweighting for upright stance between AFE≤10 and AFE>10, suggesting that soccer heading during early adolescence is not associated with balance deficits in college-aged soccer players, notwithstanding potential deficits in other markers of neurological function.
Keywords: concussion, subconcussion, balance, postural control, neurodevelopment
Introduction
In 2015, US Soccer announced initiatives aimed at improving concussion awareness and management [1]. As part of these initiatives, US Soccer eliminated soccer heading for youth players ages 10 years and younger and limited soccer heading practice for children ages 11 to 13 years [1]. These rule changes are intended to prevent concussions and to reduce repetitive hits to the head [1]. In high school soccer, 28% of concussions among boys and 26.5% of concussions among girls occur during the act of soccer heading [2], so presumably, limiting soccer heading could reduce concussion incidence. In soccer, repetitive heading has been associated with changes in neuropsychological testing and in neuroimaging, although the literature is mixed [3, 4]. However, limited empirical evidence associates soccer heading at ages 10 years and younger with medium-to-long-term behavioral deficits or concussion incidence.
Younger age of first exposure (AFE) to American football has been associated with later-life impairments in cognitive function, altered corpus callosum white matter microstructure, decreased thalamic volume, greater behavioral and mood symptoms and earlier age of neurobehavioral symptom onset in former National Football League and amateur players [5–9]. In current high school and collegiate American football players, younger AFE is not associated with deficits in behavioral, cognitive, psychological or physical performance [10–12]. Together, these studies suggest that consequences of early exposure to repetitive head impacts do not manifest until later in life or symptoms of neurological impairment progress with increased exposure to repetitive head impacts throughout and beyond collegiate play [11]. These studies were limited to American football players, and although this cohort may have the highest exposure to repetitive head impacts, other sports, such as soccer, also involve repetitive hits to the head [13, 14]. Furthermore, potentially increased risk of later life cognitive impairments and neurodegenerative diseases among former soccer players [15–20] has caused researchers and clinicians to question if soccer heading should be allowed at the youth level [21, 22].
Common clinical concussion assessments include tests of behavioral, cognitive, psychological and physical outcomes [23, 24]. Although the first three have received considerable attention in the AFE literature, physical outcomes, such as postural control, have been understudied. Humans use feedback from visual, vestibular, and proprioceptive systems to maintain upright stance. Reliance on each of these three sensory systems must be dynamically reweighted (i.e., sensory reweighting) depending on what feedback is available from each system. For example, if a person enters a dark room and he/she cannot rely on vision, he/she must then rely more on vestibular and proprioceptive feedback to maintain upright stance. Herein, we use an experimental paradigm to assess sensory reweighting. Participants experience various combinations of visual, vestibular, and proprioceptive stimuli, and we monitor how they dynamically reweight sensory information to maintain upright stance [25]. Reliance on each individual sensory system (i.e., sensory modality) depends not only on that modality’s stimulus amplitude but also on the stimulus amplitude of the other simultaneously presented modalities. Specifically, the dependence of vision on visual movement amplitude is interpreted as intra-modal reweighting; whereas the dependence of vision on vestibular or proprioceptive stimuli amplitudes is interpreted as inter-modal reweighting. For example, when the amplitude of the visual stimulus increases, visual feedback is perceived to be less reliable, and so more emphasis is placed on feedback from vestibular and proprioceptive systems [25]. When vibration is turned on, proprioception is perceived to be less reliable and so more emphasis is placed on feedback from visual and vestibular systems [25].
In other AFE studies, groups were dichotomized AFE<12 and AFE>12. Age 12 was chosen as the cutoff because ages 10–12 are considered a time period of tremendous cognitive development, including macrostructural, microstructural, neural network, and vascular maturation [26]. In the context of postural control and sensory processing, development occurs at a younger age. Even infants have postural control responses to the manipulation of a single sensory system [27–29]; however sensory reweighting develops at a later age. Specifically, intra-modal reweighting is exhibited by children as young as 4 years; inter-modal reweighting typically develops later (i.e., age 10 years) [30, 31], which further supports using age 10 as a cutoff in examining the effects of AFE to soccer heading on postural control. Therefore, the purpose of this study was to compare sensory reweighting for upright stance between current college-aged soccer players who began soccer heading at ages 10 years and younger (AFE≤10) and those who began soccer heading after age 10 years (AFE>10). Consistent with other AFE findings in college-aged cohorts, we hypothesized that there would be no differences in sensory reweighting between groups.
Materials and Methods
Participants
Thirty soccer players participated in this study (Table 1). Potential participants included both men and women between ages 18 and 30 years, who were current members of a soccer team (i.e., varsity, club, intramural). Exclusionary criteria included any head, neck, or lower extremity injury in the past six months; history of vestibular or ocular dysfunction; currently taking any medications affecting balance; history of any neurological disorders (e.g., seizure disorders); unstable cardiac or pulmonary disease; goalkeepers. The protocol used was approved by the university Institutional Review Board and all participants provided written informed consent. Furthermore, this study meets the ethical standards of the journal [32].
Table 1.
Demographic information, sport participation history, and SCAT-5 data for each group (i.e., AFE≤10 years and AFE>10 years).
| AFE≤10 years | AFE>10 years | p | Effect Size | ||
|---|---|---|---|---|---|
| Demographic Information | |||||
| Number of participants | 19 | 11 | |||
| Age (years) | 22±3 | 22±2 | 0.709 | 0.147 | |
| Height (cm) | 171±9 | 172±7 | 0.860 | 0.070 | |
| Weight (kg) | 70±11 | 69±13 | 0.752 | 0.119 | |
| Sex (n) | F | 8 | 6 | 0.707 | 0.120 |
| M | 11 | 5 | |||
| Race(n) | African American | 1 | 2 | 0.024* | 0.611 |
| Asian | 0 | 4 | |||
| Multiple Races | 1 | 0 | |||
| Not Reported | 1 | 1 | |||
| White | 16 | 4 | |||
| Ethnicity (n) | Hispanic | 4 | 1 | 0.626 | 0.155 |
| Not Hispanic | 15 | 10 | |||
| Sport Participation History | |||||
| Concussion hx (n) | 0 | 14 | 8 | 0.913 | 0.078 |
| 1 | 4 | 2 | |||
| 2 | 1 | 1 | |||
| Soccer participation (years) | 16±5 | 10±4 | 0.001* | 1.427 | |
| Current level of play (n) | Varsity | 4 | 0 | 0.238 | 0.309 |
| Club | 12 | 8 | |||
| Intramural | 3 | 3 | |||
| Symptoms after heading (n) | 2±3 | 8±6 | 0.017* | 1.123 | |
| Symptom frequency after heading | 3±5 | 12±12 | 0.006* | 1.000 | |
| SCAT-5 | |||||
| Hospitalized for head injury (n) | No | 16 | 10 | 1.000 | 0.095 |
| Yes | 3 | 1 | |||
| Migraine hx (n) | No | 18 | 10 | 1.000 | 0.074 |
| Yes | 1 | 1 | |||
| LD hx (n) | No | 19 | 11 | N/A | N/A |
| ADHD hx (n) | No | 17 | 11 | 0.520 | 0.203 |
| Yes | 2 | 0 | |||
| Depression hx (n) | No | 18 | 11 | 1.000 | 0.141 |
| Yes | 1 | 0 | |||
| Symptoms at baseline (n) | 2±3 | 4±5 | 0.149 | 0.510 | |
| Symptom severity at baseline | 3±3 | 7±10 | 0.127 | 0.526 | |
| Orientation | 5 | 5 | N/A | N/A | |
| Immediate memory | 23±3 | 22±5 | 0.640 | 0.171 | |
| Concentration | 5±1 | 4±1 | 0.609 | 0.194 | |
| BESS | 4±5 | 4±3 | 0.735 | 0.138 | |
| Delayed recall | 8±1 | 8±2 | 0.804 | 0.096 | |
Notes: AFE = age of first exposure, F = female, M = male, hx = history, LD = learning disability, ADHD = attention-deficit/hyperactivity disorder.
Experimental setup
Participants stood in the Immersive Labs (Bertec Corporation, Columbus, OH, USA) with their feet positioned such that the distance between their heels was ~11% of their height and at an angle of 14° from the midline [25]. Participants experienced four conditions: low amplitude visual scene movement - vibration - GVS; low amplitude visual scene movement - GVS - no vibration; high amplitude visual scene movement - vibration - GVS; high amplitude visual scene movement - GVS - no vibration [25]. Five trials from each condition were run for a total of twenty trials. The length of each trial was 125 seconds with 5 seconds added at the beginning and end of each trial (total 135 seconds) to allow the sensory perturbations to ramp up and ramp down [25]. The visual, vestibular, and proprioceptive sensory stimuli were similar to those used by Hwang and colleagues [25]. Different frequencies were chosen for each stimulus so that we could measure a response to each modality independently and so that they did not share common low-order harmonics [25].
Visual sensory stimulus
The visual display, consisting of 500 randomly distributed white pyramids on a black background, was projected using the Immersive Labs projector onto the dome surface. The visual signal was displayed as a visual sinusoidal translation in the anterior-posterior (AP) direction at 0.2Hz. The visual stimulus was presented at different amplitudes (i.e., 0.2m and 0.8m in Unity, Unity Technologies, San Francisco, CA) to measure the approximate center of mass (COM) AP change in gain to vision, an intra-modal effect; and change in gain to GVS and vibration, both inter-modal effects.
Proprioceptive sensory stimulus
A pair of custom vibrators were designed to apply bilateral vibration of the Achilles’ tendons. The vibrators consisted of two 20mm vibrator motors, driven at 80Hz and 1mm amplitude displacement, encased in a PVC container with a flexible surface mounted on the contact face for comfortable fitting around the Achilles’ tendons. The vibrators were attached to the leg by an elastic strap. The proprioceptive signal was applied as a square-wave periodic stimulus with equal on and off time durations at 0.28Hz.
Vestibular sensory stimulus
Binaural-monopolar galvanic vestibular stimulation (GVS) was applied via the neuroConn DC-STIMULATOR PLUS (neuroCare Group, Munich, Germany). The vestibular stimulus was delivered through circular electrodes (1.25” Round Pals Electrodes, Axelgaard Manufacturing Co Ltd, Fallbrook, CA, USA) secured to the mastoid process and to 2cm ipsilateral to the T2 spinous process. Before attaching the electrodes, conductive electrode gel was applied at the electrode-skin interface to improve conductance. The GVS signal consisted of a ±1mA sinusoidal galvanic stimulus at 0.36Hz.
Kinematics
Kinematics were recorded in QTM with 12 Miqus motion capture cameras (Qualisys Inc., Göteborg, Sweden) at 100Hz. Twelve reflective markers were placed bilaterally on the temple, acromioclavicular joint, greater trochanter, lateral femoral condyle, lateral malleolus, and first metatarsal head. The approximate COM was determined by the midpoint of left and right greater trochanter markers.
Spectral analysis
Postural responses to the sensory stimuli were measured by the frequency response function at the driving frequencies (0.2Hz for vision, 0.28Hz for vibration, 0.36Hz for GVS, Figure 1). One frequency response function was calculated for COM sway to each sensory stimulus (i.e., vision, vibration, GVS), thus a total of three frequency response functions were calculated for each trial (i.e., COM to vision, COM to vibration, COM to GVS). The frequency response function is a complex number with gain (absolute value of the frequency response function) representing the COM sway amplitude divided by the stimulus amplitude and phase (argument of the frequency response function) representing the temporal relationship between COM sway and stimulus (i.e., the COM sway may lead the stimulus (positive values) or lag behind it (negative values)). Because each stimulus was presented at a different frequency, the gain to each stimulus corresponds to the separate contribution of each stimulus to a participants’ postural response and reflects the coupling of postural sway to the stimulus motion.
Figure 1.
Gain/phase calculation and interpretation. (A) The cross spectral density (e.g., COM and visual stimulus) and the power spectral density of the stimulus were used to compute (B) the frequency response function at the driving frequencies (e.g., 0.2Hz for vision). The frequency response function is a complex number with gain representing (C) the COM A/P sway amplitude relative to the stimulus amplitude and phase representing (D) the temporal relationship between COM sway and stimulus. (A) Solid line represents the power spectral density of the COM; dashed line represents that power spectral density of the visual stimulus; dotted line represents the cross spectral density of the COM and visual stimulus. (B) Dots represent the individual frequency response functions for each trial of a condition; the star represents the average of the 5 trials. Gain is the distance from the origin, phase is the angle from 0 degrees (θ). (C, D) The dashed line represents the COM response, the solid line represents the stimulus. The relative amplitude is represented by the gain. The relative offset is represented by the phase.
The frequency response function was calculated as the cross spectral density (i.e., COM and vision, COM and vibration, COM and GVS) divided by the power spectral density of the stimulus (i.e., vision, vibration, GVS). The cross spectral density of two signals is the equivalent of the cross-correlation in the frequency domain and provides the power shared at a given frequency for the two signals. The power spectral density corresponds to the power of the stimulus across all frequencies. Both the cross spectral density and the power spectral density were computed using Welch’s method with a 50s Hanning window and 50% overlap and then averaged across trials.
Time domain analysis
The COM 95% area and COM sway velocity were calculated from the COM AP and ML displacement. The COM 95% area was calculated by drawing an ellipse around intersecting lines that describe the maximum AP and ML sway and quantifying the area of the ellipse [33]. The COM sway velocity was calculated by taking the total distance traveled and dividing it by the length of time (125 s) of the trial [33].
Questionnaires and SCAT-5
In addition to the sensory reweighting protocol described above, participants also completed a custom demographic and sport history questionnaire (DASQ, Appendix 1) and the Sport Concussion Assessment Tool 5th Edition (SCAT-5). The DASQ captured demographic information, concussion history, and soccer participation history, including age of first exposure to soccer heading. Participants also reported symptoms experienced following soccer heading. Specifically, the question read, “Usually heading is unremarkable. Sometimes, though, heading is not quite right and may cause dizziness, confusion, or other feelings. Please indicate how often you experience the following symptoms of heading.” Participants then rated symptoms (i.e., taken from the SCAT-5 symptom checklist) on a Likert scale from 0 to 4, where 0 was “Never”, 1 was “Rarely (1–25% of the time),” 2 was “Sometimes (26–50% of the time),” 3 was “Often (50–75% of the time),” and 4 was “Very Often (76–100% of the time).” The SCAT-5 is a brief assessment tool designed and used for evaluating concussions, which includes symptom evaluation, and tests of orientation, immediate memory, concentration, balance, and delayed recall [34].
Statistical analysis
We compared demographic information, concussion history, soccer participation history, and SCAT-5 data using independent samples t-tests for continuous variables and chi-squared for count variables. To compare sensory reweighting between groups, we used a repeated measures ANOVA. The within-subjects effects included change in visual amplitude and effect due to vibration on/off. The between-group effect included group (i.e., AFE≤10 years, AFE>10 years). Because years of sport participation were different between groups, we first ran the analyses unadjusted for years of sport participation and then ran the analyses adjusted for years of sport participation. There were eight outcomes analyzed independently, including gain and phase of COM relative to the visual stimulus, to the GVS stimulus, and to the vibration stimulus, COM 95% area, and COM sway velocity. The gain/phase to vibration can be measured only with vibration turned on (low amplitude visual scene movement - vibration - GVS and high amplitude visual scene movement - vibration – GVS conditions). Significance was defined a priori at p<0.05.
Results
Questionnaires and SCAT-5
There were no differences in sex, age, height, or weight between groups (Table 1). Concussion history also did not differ between groups, although years of soccer participation did differ between groups, whereby AFE≤10 years played soccer for more years than AFE>10 years (Table 1, p=0.001, effect size=1.427). Surprisingly, AFE>10 years reported a greater number of symptoms associated with soccer heading (Table 1, p=0.017, effect size=1.123) and greater frequency (i.e., ∑(symptoms*frequency); Table 1, p=0.006, effect size=1.000) than AFE≤10 years. There were no differences in SCAT-5 outcome measures (Table 1).
Sensory reweighting
Gains to each modality across conditions are presented in Figure 2. Phases are presented in Figure 3. Results of the repeated measures ANOVA suggested that there were no differences (i.e., unadjusted and adjusted models) in gains to any modality or phases between groups (Table 2, Table 3). Additionally, years of soccer participation was not associated with gains to any modality or phases (Table 2, Table 3). Changes in visual amplitude and vibration on/off were associated with changes in gains to each modality across conditions (Table 2), but not in phases across conditions (Table 3).
Figure 2:
Gain to each modality. The grey lines represent the group self-reporting AFE≤10 years. The black lines represent the group self-reporting AFE>10 years. Error bars represent standard error. There are no significant differences between groups.
Figure 3:
Phase of each modality. The grey lines represent the group self-reporting AFE≤10 years. The black lines represent the group self-reporting AFE>10 years. Error bars represent standard error. There are no significant differences between groups.
Table 2.
Results of the repeated measures ANOVA for gain to each modality.
| Gain to GVS | Gain to Vibration | Gain to Vision | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Unadjusted Model | F1,28 | p | Partial η2 | F1,28 | p | Partial η2 | F1,28 | p | Partial η2 |
| Vision | 1.452 | .238 | .049 | 8.318 | 007* | .229 | 105.667 | .000* | .791 |
| Vision * AFE | .049 | .827 | .002 | .068 | .797 | .002 | .116 | .736 | .004 |
| Vibration | 15.571 | .000* | .357 | 4.886 | .035* | .149 | |||
| Vibration * AFE | .068 | .796 | .002 | .291 | .594 | .010 | |||
| Vision * Vibration | .686 | .414 | .024 | 6.724 | .015* | .194 | |||
| Vision * Vibration * AFE | .000 | .999 | .000 | .216 | .645 | .008 | |||
| AFE | .001 | .971 | .000 | .001 | .974 | .000 | .033 | .857 | .001 |
| Adjusted Model | F1,27 | p | Partial η2 | F1,27 | p | Partial η2 | F1,27 | p | Partial η2 |
| Vision | .055 | .816 | .002 | 4.471 | 044* | .142 | 27.949 | <.001* | .509 |
| Vision * Years | .028 | .869 | .001 | 1.487 | .233 | .052 | 3.395 | .076 | .112 |
| Vision * AFE | .007 | .936 | .000 | .240 | .628 | .009 | .599 | .446 | .022 |
| Vibration | .189 | .667 | .007 | 1.299 | .264 | .046 | |||
| Vibration * Years | .848 | .365 | .030 | .196 | .661 | .007 | |||
| Vibration * AFE | .101 | .753 | .004 | .032 | .860 | .001 | |||
| Vision * Vibration | .009 | .927 | .000 | .732 | .400 | .026 | |||
| Vision * Vibration * Years | .148 | .703 | .005 | .000 | .990 | .000 | |||
| Vision * Vibration * AFE | .050 | .825 | .002 | .145 | .707 | .005 | |||
| Years | .558 | .461 | .020 | .137 | .715 | .005 | 1.938 | .175 | .067 |
| AFE | .162 | .691 | .006 | .057 | .812 | .002 | .911 | .348 | .033 |
Notes: The * indicates statistical significance, AFE = between group effect (i.e., AFE<10 years vs. AFE>10 years), Vision = effect due to change in visual amplitude, Vibration = effect due to vibration on/off, Years = effect due to years of soccer participation.
Table 3.
Results of the repeated measures ANOVA for phase of each modality.
| GVS Phase | Vibration Phase | Vision Phase | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Unadjusted Model | F1,28 | p | Partial η2 | F1,28 | p | Partial η2 | F1,28 | p | Partial η2 |
| Vision | 7.946 | .009* | .221 | .562 | .460 | .020 | .748 | .394 | .026 |
| Vision * AFE | .893 | .353 | .031 | .068 | .796 | .002 | 1.224 | .278 | .042 |
| Vibration | .045 | .833 | .002 | .627 | .435 | .022 | |||
| Vibration * AFE | .758 | .391 | .026 | 2.803 | .105 | .091 | |||
| Vision * Vibration | 2.446 | .129 | .080 | .801 | .379 | .028 | |||
| Vision * Vibration * AFE | .632 | .433 | .022 | 1.191 | .285 | .041 | |||
| AFE | .295 | .591 | .010 | .327 | .572 | .012 | 2.267 | .143 | .075 |
| Adjusted Model | F1,27 | p | Partial η2 | F1,27 | p | Partial η2 | F1,27 | p | Partial η2 |
| Vision | 4.126 | .052 | .133 | .108 | .744 | .004 | .153 | .699 | .006 |
| Vision * Years | 1.339 | .257 | .047 | .008 | .929 | .000 | .013 | .909 | .000 |
| Vision * AFE | .012 | .914 | .000 | .025 | .876 | .001 | .673 | .419 | .024 |
| Vibration | .000 | .985 | .000 | .000 | .987 | .000 | |||
| Vibration * Years | .003 | .958 | .000 | .065 | .800 | .002 | |||
| Vibration * AFE | .531 | .472 | .019 | 2.225 | .147 | .076 | |||
| Vision * Vibration | 1.683 | .205 | .059 | .278 | .602 | .010 | |||
| Vision * Vibration * Years | .687 | .414 | .025 | .063 | .804 | .002 | |||
| Vision * Vibration * AFE | 1.264 | .271 | .045 | .535 | .471 | .019 | |||
| Years | 2.757 | .108 | .093 | 1.511 | .230 | .053 | .292 | .593 | .011 |
| AFE | .251 | .620 | .009 | .057 | .813 | .002 | 2.328 | .139 | .079 |
Notes: The * indicates statistical significance, AFE = between group effect (i.e., AFE<10 years vs. AFE>10 years), Years = effect due to years of soccer participation, Vision = effect due to change in visual amplitude, Vibration = effect due to vibration on/off.
The COM 95% area and COM sway velocity are presented in Figure 4. Results of the repeated measures ANOVA suggested that there were no differences (i.e., unadjusted and adjusted models in COM 95% area or COM sway velocity between groups (Table 4). Additionally, years of soccer participation was not associated with COM 95% area or COM sway velocity (Table 4). Changes in visual amplitude and vibration on/off were not associated with changes in COM 95% area or COM sway velocity across conditions (Table 4).
Figure 4.
COM 95% area and COM sway velocity across conditions. The grey lines represent the group self-reporting AFE≤10 years. The black lines represent the group self-reporting AFE>10 years. Error bars represent standard error. There are no significant differences between groups.
Table 4.
Results of the repeated measures ANOVA for COM 95% area and COM sway velocity.
| 95% Area | Sway Velocity | |||||
|---|---|---|---|---|---|---|
| Unadjusted Model | F1,28 | p | Partial η2 | F1,28 | p | Partial η2 |
| Vision | .475 | .496 | .017 | 1.334 | .258 | .046 |
| Vision * AFE | .000 | .994 | .000 | .197 | .661 | .007 |
| Vibration | 10.557 | .003* | .274 | 1.480 | .234 | .050 |
| Vibration * AFE | .059 | .811 | .002 | .075 | .786 | .003 |
| Vision * Vibration | .936 | .342 | .032 | 6.243 | .019* | .182 |
| Vision * Vibration * AFE | .365 | .550 | .013 | .735 | .399 | .026 |
| AFE | .786 | .383 | .027 | .673 | .419 | .023 |
| Adjusted Model | F1,27 | p | Partial η2 | F1,27 | p | Partial η2 |
| Vision | .028 | .868 | .001 | .305 | .585 | .011 |
| Vision * Years | .004 | .953 | .000 | .035 | .853 | .001 |
| Vision * AFE | .001 | .977 | .000 | .215 | .647 | .008 |
| Vibration | 2.018 | .167 | .070 | .305 | .585 | .011 |
| Vibration * Years | .146 | .705 | .005 | .027 | .870 | .001 |
| Vibration * AFE | .001 | .979 | .000 | .100 | .755 | .004 |
| Vision * Vibration | 1.531 | .227 | .054 | .012 | .912 | .000 |
| Vision * Vibration * Years | .945 | .340 | .034 | .566 | .458 | .021 |
| Vision * Vibration * AFE | .005 | .946 | .000 | 1.275 | .269 | .045 |
| Years | .467 | .500 | .467 | .364 | .551 | .013 |
| AFE | .104 | .749 | .104 | .099 | .756 | .004 |
Notes: The * indicates statistical significance, AFE = between group effect (i.e., AFE<10 years vs. AFE>10 years), Years = effect due to years of soccer participation, Vision = effect due to change in visual amplitude, Vibration = effect due to vibration on/off.
Discussion
Despite recent rule changes regarding soccer heading among youth athletes, limited empirical evidence associates soccer heading at ages 10 years and younger with medium-to-long-term behavioral deficits. Therefore, the purpose of this study was to compare sensory reweighting for upright stance between college-aged soccer players who began soccer heading ages 10 years and younger and those who began soccer heading after age 10 years. Groups were similar in demographic characteristics and concussion history, although AFE≤10 years played soccer for longer (Table 1). As we hypothesized, there were no differences in sensory reweighting between groups (Figures 2–4; Tables 2–4). This suggests that soccer heading at ages 10 years and younger is not associated with balance deficits in college-aged soccer players, notwithstanding potential deficits in other markers of neurological function.
In 2015, Stamm and colleagues examined the association between AFE to tackle football and later-life cognitive function in 42 former NFL players [8]. They divided participants into 2 groups based on AFE to tackle football (i.e., AFE<12 years, AFE≥12 years) and reported that AFE<12 years exhibited significantly worse performance on the Wisconsin Card Sorting Test, the Neuropsychological Assessment Battery List Learning and the Wide Range Achievement Test, 4th Edition [8]. This group followed up their initial work with subsequent studies, including neuroimaging and neuropsychiatric assessments, confirming that earlier AFE to tackle football may contribute to later-life neurological impairments in former NFL players [5–9]. In 2016, Solomon and colleagues failed to replicate the findings of Stamm and colleagues; in 45 retired NFL athletes, there were no associations between years of exposure to pre-high school football and neuroradiological, neurological, and neuropsychological outcome measures [35]. In 2019, a series of studies in current high school and collegiate athletes reported no associations between estimated AFE to tackle football or other contact sports and baseline behavioral, cognitive, psychological or physical performance [10–12]. Because the AFE work to date is largely limited to American football players, the effects of soccer heading during early adolescence on later life brain health is unknown. The current study adds to the growing body of literature suggesting that AFE to contact sports is not associated with worse neurological performance (e.g., on postural control assessments) in current college-aged athletes.
The initial hypothesis posed by Stamm and colleagues was that exposure to repetitive head impacts during youth may interfere with healthy neurodevelopmental maturation, increasing vulnerability to neurological consequences [8]. Because sensory reweighting for upright stance develops at a younger age than some cognitive processes, our cohorts were divided differently than Stamm and colleagues (i.e., AFE≤10, AFE>10). Although, we cannot discern the effects of AFE on later-life postural control impairments, our findings lend empirical evidence to the notion that younger AFE is not associated with postural control. Conflicting findings between former American football players and current contact sport athletes may suggest that (1) consequences of early exposure to repetitive head impacts do not manifest until later in life, (2) symptoms of neurological impairment progress with increased exposure to repetitive head impacts throughout and beyond collegiate play, or (3) improvements in concussion management over the past decade has reduced long-term deficits in neurological function. To test these possibilities, we must continue to observe these athletes across the lifespan.
Surprisingly, AFE>10 years reported a greater number of symptoms associated with soccer heading and greater frequency than AFE≤10 years. Several possibilities may explain these findings: (1) athletes who began soccer heading at a younger age are more skilled and therefore, may have better technique when heading the ball [36–38]; (2) players who were afraid to head the ball began soccer heading at a later age and therefore, are more likely to experience symptoms when heading; (3) players who began soccer heading at a later age are more symptomatic at baseline (Table 1). At baseline, AFE≤10 years reported 2±3 symptoms and 3±3 symptom severity, while AFE>10 years reported 4±5 symptoms and 7±10 symptom severity. Although this comparison did not reach statistical significance, AFE>10 years reported a greater number of symptoms and greater symptom severity at baseline than AFE≤10 years. The symptoms reported after soccer heading among the AFE>10 years cohort may be a manifestation of their baseline symptom presentation exacerbated with soccer heading. Future research should aim to replicate these findings in a larger cohort and to compare symptom presentation at baseline between athletes who began sport participation at a young age and those who began sport participation at an older age. It is possible that increased physical activity at a young age results in a decreased symptom burden later in life (e.g., in college-aged athletes).
In soccer, over 25% of concussions occur during the act of heading [2]. It is reasonable to believe that athletes who began soccer heading at a younger age may be more likely to sustain concussions. However, this was not the case. Instead, there was no difference in concussion history between groups. It should be noted that although over 25% of concussions occur during the act of heading, most concussions are attributed to non-heading related events [2]. This is a retrospective, cohort study and was not designed to prospectively address the question, “does limiting soccer heading reduce concussion incidence?” We also did not compare age of first concussion because of low sample sizes among our concussion cohorts. Schmidt and colleagues reported that participants self-reporting their first concussion during childhood had an increased risk of sustaining subsequent concussions (RR=2.19) compared to participants self-reporting their first concussion during adolescence [39]. Future research should examine the association between age of first concussion and later-life outcomes. Ultimately, our findings suggest that beginning soccer heading at ages 10 years or younger does not result in higher concussion incidence among college-aged soccer players.
The change in visual amplitude and effect due to vibration on/off had significant effects on gains to each sensory modality (Table 2). Gain to vision decreased from the low vision condition to the high vision condition, as expected (Figure 2). During the high amplitude visual stimulus, vision is perceived to be less reliable (i.e., gain to vision decreases) and so more emphasis is placed on feedback from vestibular and proprioceptive systems (i.e., gain to GVS and gain to vibration increase, Figure 2) [25]. When vibration is turned on, proprioception is perceived to be less reliable (i.e., gain to vibration decreases) and so more emphasis is placed on feedback from visual and vestibular systems (i.e., gain to vision and gain to GVS increase, Figure 2) [25]. Phase, on the other hand, stays relatively consistent across conditions (Table 3, Figure 3) [25]. There were absolute differences in phases across modalities (Figure 3), which may be attributed to known differences among time delays for different modalities. For example, vision had the largest delay lag (approximately −30–0 degrees), while GVS and vibration both had phase leads (approximately 0–60 degrees) [25].
This study was the first to examine AFE in a cohort of college-aged soccer players and the first to report on postural control, specifically sensory reweighting for upright stance. However, this study is not without limitations. First, we included a limited sample size of college-aged soccer players. Future research should extend these findings in other contact sports, such as ice hockey and lacrosse, and across the lifespan. Second, due to the small sample size, we had to dichotomize AFE (i.e., AFE≤10 and AFE>10). This assumes that our cutoff age of 10 years is the most relevant to the aims addressed herein. With a larger sample size, we can examine AFE as a continuous variable to determine if there are other ages that may be more appropriate cutoffs. Finally, we only addressed sensory reweighting for upright stance. Postural control assessments are frequently used in concussion assessments [40], yet future research should include behavioral, cognitive and psychological assessments.
The purpose of this study was to compare sensory reweighting for upright stance between college-aged soccer players based on AFE to soccer heading (i.e., AFE≤10 and AFE>10). As we hypothesized, there were no differences in sensory reweighting for upright stance between AFE≤10 and AFE>10. These findings suggest that soccer heading during early adolescence is not associated with balance deficits in college-aged soccer players, notwithstanding potential deficits in other markers of neurological function or potential deficits that develop later in life.
Acknowledgements:
This study was supported by the NIH-NINDS R01 (NS102157-01) grant, “Behavioral- and bio-markers of subconcussion with controlled human head impact.”
Appendix 1. Demographic and Sport History Questionnaire
References
- 1.Yang YT, Baugh CM. US youth soccer concussion policy: Heading in the right direction. JAMA Pediatr. 2016;170(5):413–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Kerr ZY, Campbell KR, Fraser MA, et al. Head impact locations in US high school boys’ and girls’ soccer concussions, 2012/13–2015/16. J Neurotrauma. 2019;36(13):2073–2082. [DOI] [PubMed] [Google Scholar]
- 3.Kontos AP, Braithwaite R, Chrisman SP, et al. Systematic review and meta-analysis of the effects of football heading. Br J Sports Med. 2017;51:1118–1124. [DOI] [PubMed] [Google Scholar]
- 4.Maher ME, Hutchison M, Cusimano M, et al. Concussions and heading in soccer: A review of the evidence of incidence, mechanisms, biomarkers and neurocognitive outcomes. Brain Inj. 2014;28(3):271–285. [DOI] [PubMed] [Google Scholar]
- 5.Stamm JM, Koerte IK, Muehlmann M, et al. Age at first exposure to football is associated with altered corpus callosum white matter microstructure in former professional football players. J Neurotrauma. 2015;32(22):1768–1776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schultz V, Stern RA, Tripodis Y, et al. Age at first exposure to repetitive head impacts is associated with smaller thalamic volumes in former professional American football players. J Neurotrauma. 2018;35(2):278–285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Alosco ML, Kasimis AB, Stamm JM, et al. Age of first exposure to American football and long-term neuropsychiatric and cognitive outcomes. Transl Psychiatry. 2017;7:e1236. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stamm JM, Bourlas AP, Baugh CM, et al. Age of first exposure to football and later-life cognitive impairment in former NFL players. Neurology. 2015;84(11):1114–1120. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Alosco ML, Mez J, Tripodis Y, et al. Age of first exposure to tackle football and chronic traumatic encephalopathy. Ann Neurol. 2018;83(5):886–901. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Caccese JB, Iverson GL, Cameron KL, et al. Estimated Age of First Exposure to Contact Sports is Not Associated with Greater Symptoms or Worse Cognitive Functioning in U.S. Service Academy Athletes. J Neurotrauma. 2019. [Epub ahead of print]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Caccese JB, DeWolf RM, Kaminski TW, et al. Estimated age of first exposure to american football and neurocognitive performance amongst NCAA male student-athletes: A cohort study. Sports Med. 2019;49(3):477–487. [DOI] [PubMed] [Google Scholar]
- 12.Brett BL, Huber DL, Wild A, et al. Age of first exposure to american football and behavioral, cognitive, psychological, and physical outcomes in high school and collegiate football players. Sports Health. 2019. [Epub ahead of print]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Caccese JB, Lamond LC, Buckley TA, Kaminski TW. Reducing purposeful headers from goal kicks and punts may reduce cumulative exposure to head acceleration. Res Sports Med. 2016;24(4):407–415. [DOI] [PubMed] [Google Scholar]
- 14.Lamond LC, Caccese JB, Buckley TA, et al. Linear acceleration in direct head contact across impact type, player position, and playing scenario in collegiate women’s soccer players. J Athl Train. 2018;53(2):115–121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Mackay DF, Russell ER, Stewart K, et al. Neurodegenerative disease mortality among former professional soccer players. N Engl J Med. 2019;381:1801–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lipton ML, Kim N, Zimmerman ME, et al. Soccer heading is associated with white matter microstructural and cognitive abnormalities. Radiology. 2013;268(3):850–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Downs DS, Abwender D. Neuropsychological impairment in soccer athletes. J Sport Med Phys Fit. 2002;42(1):103–107. [PubMed] [Google Scholar]
- 18.Straume-Naesheim TM, Andersen TE, Dvorak J, Bahr R. Effects of heading exposure and previous concussions on neuropsychological performance among Norwegian elite footballers. Br J Sports Med. 2005;39(suppl 1):i70–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tysvaer AT, Løchen EA. Soccer injuries to the brain: A neuropsychologic study of former soccer players. Am J Sports Med. 1991;19(1):56–60. [DOI] [PubMed] [Google Scholar]
- 20.Sortland O, Tysvaer AT. Brain damage in former association football players. Neuroradiology. 1989;31(1):44–8. [DOI] [PubMed] [Google Scholar]
- 21.O’Kane JW. Is heading in youth soccer dangerous play? Phys Sportsmed. 2016;44(2):190–4. [DOI] [PubMed] [Google Scholar]
- 22.Putukian M. Heading in soccer: is it safe? Curr Sports Med Rep. 2004;3(1):9–14. [DOI] [PubMed] [Google Scholar]
- 23.Broglio SP, McCrea M, McAllister T, et al. A national study on the effects of concussion in collegiate athletes and US military service academy members: The NCAA–DoD concussion assessment, research and education (CARE) consortium structure and methods. Sports Med. 2017;47(7):1437–1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Broglio SP, Kontos AP, Levin H, et al. The National Institute of Neurological Disorders and Stroke and Department of Defense sport-related concussion common data elements version 1.0 recommendations. J Neurotrauma. 2018;35(23):2776–2783. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hwang S, Agada P, Kiemel T, Jeka JJ. Dynamic reweighting of three modalities for sensor fusion. PloS One. 2014;9(1):e88132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Alosco ML, Stern RA. Youth exposure to repetitive head impacts from tackle football and long-term neurologic outcomes: A review of the literature, knowledge gaps and future directions, and societal and clinical implications. Semin Pediatr Neurol. 2019. [Epub ahead of print]. [DOI] [PubMed] [Google Scholar]
- 27.Foster EC, Sveistrup H, Woollacott MH. Transitions in visual proprioception: A cross-sectional developmental study of the effect of visual flow on postural control. J Mot Behav. 1996;28(2):101–112. [DOI] [PubMed] [Google Scholar]
- 28.Barela J, Jeka JJ, Clark JE. The use of somatosensory information during the acquisition of independent upright stance. Infant Behav Dev. 1999;22(1):87–102. [Google Scholar]
- 29.Barela JA, Jeka JJ, Clark JE. Postural control in children. Exp Brain Res. 2003;150(4):434–442. [DOI] [PubMed] [Google Scholar]
- 30.Bair W, Kiemel T, Jeka JJ, Clark JE. Development of multisensory reweighting for posture control in children. Exp Brain Res. 2007;183(4):435–446. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Forssberg H, Nashner LM. Ontogenetic development of postural control in man: Adaptation to altered support and visual conditions during stance. J Neurosci. 1982;2(5):545–552. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Harriss DJ, Macsween A, Atkinson G. Ethical Standards in Sport and Exercise Science Research: 2020 Update. Int J Sports Med. 2019; 40: 813–817. [DOI] [PubMed] [Google Scholar]
- 33.Prieto TE, Myklebust JB, Hoffmann RG, et al. Measures of postural steadiness: Differences between healthy young and elderly adults. IEEE Trans Biomed Eng. 1996;43(9):956–966. [DOI] [PubMed] [Google Scholar]
- 34.Echemendia RJ, Meeuwisse W, McCrory P, et al. The sport concussion assessment tool 5th edition (SCAT5): Background and rationale. Br J Sports Med. 2017;51(11):848–850. [DOI] [PubMed] [Google Scholar]
- 35.Solomon GS, Kuhn AW, Zuckerman SL, et al. Participation in Pre–High school football and neurological, neuroradiological, and neuropsychological findings in later life: A study of 45 retired National Football League players. Am J Sports Med. 2016;44(5):1106–1115. [DOI] [PubMed] [Google Scholar]
- 36.Caccese JB, Kaminski TW. Minimizing head acceleration in soccer: A review of the literature. Sports Med. 2016;46(11):1591–1604. [DOI] [PubMed] [Google Scholar]
- 37.Caccese JB, Buckley TA, Tierney RT, et al. Sex and age differences in head acceleration during purposeful soccer heading. Res Sports Med. 2018;26(1):64–74. [DOI] [PubMed] [Google Scholar]
- 38.Caccese JB, Buckley TA, Tierney RT, et al. Head and neck size and neck strength predict linear and rotational acceleration during purposeful soccer heading. Sports Biomechanics. 2018;17(4):462–476. [DOI] [PubMed] [Google Scholar]
- 39.Schmidt JD, Rizzone K, Hoffman NL, et al. Age at first concussion influences the number of subsequent concussions. Pediatr Neurol. 2018;81:19–24. [DOI] [PubMed] [Google Scholar]
- 40.Buckley TA, Oldham JR, Caccese JB. Postural control deficits identify lingering post-concussion neurological deficits. J Sport Health Sci. 2016;5(1):61–69. [DOI] [PMC free article] [PubMed] [Google Scholar]