There is increasing awareness and concern about the long-term health consequences of cumulative brain trauma in sports. A concussion or other brain injury can occur in any organized or unorganized sport; however, they are most commonly associated with contact-collision sports such as football, ice and field hockey, lacrosse, wrestling, soccer, and basketball. In particular, football predisposes players to numerous impact forces that may cause the brain to become injured through direct or indirect contact. Among boys, more concussions are attributable to football than any other sport (19). Furthermore, football is the leading cause of child and adolescent emergency room visits resulting from organized team sports activities (5).
Much of the recent attention surrounding brain injury and football has focused on current and former professional players from the National Football League (NFL). Whereas the media-led spotlight on such high-profile players is not surprising, it is important to recognize that most football players are younger, amateur athletes. In fact, approximately 70% of the estimated five million contact football players in the United States are below high school age. There are additional implications to the participation numbers, considering the effects of concussion are often more severe in this population and these athletes have a longer window of subsequent risk exposure after an initial concussion (21). Furthermore, the concern for young players is not limited to concussive injuries but also encompasses the potential cumulative damage to the developing brain caused by repetitive subconcussive impact exposure.
Head impact exposure (HIE) has been measured extensively in high school and collegiate football players, including several multiteam and multiyear studies. Throughout these seminal studies, more than 1.5 million head impacts have been captured (14). However, limited data exist pertaining to the HIE of youth football players. To date, only five studies have reported HIE in youth football players younger than high school age. Players 7 to 8 yr old were evaluated for one season by Daniel et al. and over two seasons by Young et al. The players involved in those studies only participated in nine practices and five games, and 10 practices and six games, respectively. Cobb et al. monitored 50 players (9–12 yr old) from three teams over an entire season (∼22 sessions per player); and most recently, Daniel et al. studied 17 middle school players (12–14 yr old) who participated, on average, in 29 practices and nine games in one season. Wong et al. tracked 22 players (12–13 yr old) for an entire season using a commercial impact sensor, but those data cannot be directly compared to the other studies because of differences in methodology. Although this research has provided an initial glimpse of the HIE of youth football players younger than high school age, the data are clearly not extensive enough to draw meaningful conclusions or generalizations for all of youth football, nor do they relate HIE to any measure of neurologic function.
Most studies on HIE in football have relied on the accelerometry-based head impact telemetry (HIT) system (Simbex, Lebanon, NH). These investigations have primarily quantified the frequency and magnitude of head impacts and characterized those head impacts by helmet location and playing position. In general, there are an increased number and magnitude of head impacts at older levels of play (progressing from youth to high school to college). Still, youth football players are exposed to head impacts similar to the high-magnitude impacts experienced among college players, although they occur at a lower frequency (10). A more in-depth analysis of how HIE in youth football players compares with other age groups is challenging, as existing head impact data for these individuals are scarce and are currently insufficient for accurate representation of this population.
It is not necessary for youth athletes to have sustained a concussion to have incurred a brain injury. There is growing evidence of neurologic impairment and/or neurophysiologic changes to the brain of nonconcussed football players (2,32). However, the specific factors responsible for these deficits have not been clearly established. We recently studied the effect of a full football season on neurologic function in youth football players and did not find any functional impairment using standard clinical evaluation methods (13). However, we did not have head impact data to probe for potential associations between HIE and individual changes in neurologic function.
Taken together, there is an absence of substantive information on head impact data in youth football players, either to describe HIE or to associate HIE with possible impairments in neurologic function. Our aim in the current study was to quantify and characterize HIE in adolescent football players and examine potential associations between HIE and acute impairments in neurologic function. We hypothesized that HIE of the adolescent football players in our study would be intermediate to what has been reported for high school players and younger youth players. We further hypothesized that short-term individual deficits in neurologic function, if present, would be associated with the HIE of those players. Examining possible links between repetitive brain trauma and neurologic impairment, even in the absence of a diagnosed concussion, is important for clarifying the potential risk of brain injury that all young football players encounter.
Twenty-two adolescent boys (12.9 ± 0.6 yr) from the same youth football team participated in this study. Evaluations of neurologic function consisting of self-reported symptom scores and tests of postural stability, oculomotor performance, computer-based neurocognitive testing, and reaction time (explained further below) were performed before and after a 12-wk season (27 practices and nine games). Preseason (PRE) testing was carried out during the first week of practice but before any contact, whereas postseason (POST) testing was conducted within 1 wk of the last game, approximately 13 weeks after PRE. All tests evaluated at POST were performed in the same manner as PRE by the same group of trained investigators. A subset of the assessments was also used to evaluate the players immediately after the seventh game of the season (POST-G). For any players experiencing a concussion, PRE assessments were repeated at each follow-up visit to our sports medicine clinic. Head impact exposure was measured with the HIT system at each practice and game throughout the season.
One player was unable to complete the season owing to injury (nonconcussive) and only participated in three sessions (practices). Another player quit the team for unknown reasons after participating in 11 practices and two games. An injury (nonconcussive) during the first practice caused a third player to miss most of the season, although he did return for the last practice and played in the final game (participated in a total of three sessions). A fourth player had a concussion during the seventh game of the season and missed the remaining two games and five practices. The remaining players were active throughout the entire season, although overall attendance varied. Data from all players are included in overall impact frequency, magnitude, and location distribution summaries. Individual head impact frequency for the entire season is reported based on participation, whereas the analysis of HIE versus changes in PRE to POST/POST-G neurologic function is limited to those players who completed the respective neurologic assessments. The Institutional Review Board of Sanford Research/USD, Sanford Health, Sioux Falls, SD, approved this study. All subjects gave their written assent to participate, and their parents/guardians provided written informed consent.
Health history and injury surveillance
Medical/health history and concussion recall questionnaires were administered to subjects and their parents/guardians at PRE to retrospectively assess relevant injuries, particularly head trauma and concussions, before the beginning of the season. An investigator was present during each practice and game to document injuries, and there was regular communication with the team’s coach for up-to-date injury reports throughout the season.
Testing for postural stability was performed on three separate devices: an AMTI strain gauge force platform (FP; OR6-6, Watertown, MA), a Wii fit balance board (WBB), and a NeuroCom Balance Master (NC; Clackamas, OR). Preseason and POST testing used all three devices, whereas only WBB was used for the POST-G assessment. Identical protocols were used on the FP and WBB. On these devices, the subjects performed three trials of four separate static balance tests. Each trial with the FP and WBB was conducted for 20 s at a sampling rate of 100 Hz. For each test, subjects stood with both feet shoulder-width apart on the FP or WBB while attempting to be as steady as possible. The subjects performed the tests without shoes, kept their heads in the neutral position, and had their hands on their hips. The four balance trials were the following: A) eyes open, B) eyes closed, C) eyes open while reciting the months of the year backward, and D) eyes closed while reciting the months of the year backward.
The subjects’ center of pressure (COP) positional changes on the FP were tracked and recorded using an integrated computer and diagnostic software package (BioAnalysis, AMTI, Watertown, MA). Specifically, a subject’s 95% ellipse area was obtained from BioAnalysis on each trial and used as an indicator of postural stability. This value is a measure of the elliptical area that contains 95% of the horizontal positional coordinates of a subject’s COP. In effect, the 95% ellipse area measures a subject’s postural sway over a period of time. We have shown good (intraclass correlation coefficient, 0.53–0.63) test-retest reliability measures of balance with this method in adolescents who have not had a concussion (35). Postural stability on the WBB was assessed in a similar manner, using custom software that used the raw positional coordinates of the WBB streamed via Bluetooth to calculate 95% ellipse area.
The Sensory Organization Test (SOT) was used for measuring postural stability on the NC, a device with a standing platform and a shield apparatus in front of the platform that can be programmed to tilt to alter the subject’s support surface and/or depth perception. The SOT protocol consists of six bipedal balance tests: 1) eyes open, stable surface, and stable depth perception; 2) eyes closed and stable surface; 3) eyes open, stable surface, and altered depth perception; 4) eyes open, unstable surface, and stable depth perception; 5) eyes closed and unstable surface; and 6) eyes open, unstable surface, and unstable depth perception. Three trials were performed for each test, with each trial lasting 20 s. During each test, the subjects attempted to be as steady as possible while standing with both feet shoulder-width apart directly on the force platform. The subjects performed the tests without shoes, kept their heads in the neutral position, and had their hands on their hips. Postural stability was determined from COP displacement and reported on a scale of 0–100 by the NeuroCom integrated software program.
Visual tracking and saccadic eye movements at PRE, POST, and POST-G were assessed with the King-Devick (KD) test (29). The KD test requires subjects to rapidly read single-digit numbers off a series of three cards (KD-1, KD-2, and KD-3). The arrangement and spacing of numbers on each card are unique and progressively increase in reading difficulty on each successive card. Subjects were given standard instructions and were asked to read the numbers from left to right and from top to bottom as rapidly as possible without making an error. The time to complete each card was measured with a handheld stopwatch to the nearest hundredth of a second, and the number of errors was noted. The subjects were given up to three attempts to complete each card without an error. Giving the subjects multiple attempts allowed us to obtain an accurate representation of the subjects’ maximum oculomotor performance on this test that was not inhibited by a single stumble or error. The fastest time without an error was recorded for each card, and those times were summed for a best total time on all three cards (KD-T). Although reliability measures for the KD test are not available for youth the same age as our study participants, test-retest reliability has been reported as being very high among nonconcussed adults, with an intraclass correlation coefficient of 0.97 (95% confidence interval, 0.90–1.0) (17), although a mild learning effect has also been observed (18).
Computer-based neurocognitive testing was conducted at PRE using the ImPACT program (online version), a widely used commercially available software package for concussion assessment (24). ImPACT consists of a test battery measuring various domains of neurocognitive function: reaction time (RT), visual motor speed (VMS), verbal memory (VeM), and visual memory (ViM). An absolute composite score is generated for each of these neurocognitive domains and can be compared with normative data. More detailed information about ImPACT can be found in Schatz et al. (30). Intraclass correlation coefficients for test-retest reliability of the online version of ImPACT performed by high school athletes have been reported as 0.76 for RT, 0.85 for VMS, 0.62 for VeM, and 0.70 for ViM (16), although similar measures have not been reported for individuals the same age as our study population. Each PRE ImPACT evaluation was screened for invalid effort using the automated internal checks available with the software.
Simple RT (SRT) measurements were obtained at PRE, POST, and POST-G using a visual response task. The test was performed at a computer workstation using a custom PC program. The subjects were instructed to press the spacebar of the computer keyboard as quickly as possible after seeing a visual symbol appear on the video monitor. Five visual cues appeared at random time intervals, and the average time between the appearance of each cue and the depression of the spacebar was recorded by the software program as the measure of SRT.
Self-reported symptom scores were obtained at PRE, POST, and POST-G by implementing the Post-Concussion Symptom Scale (PCSS) (23). This symptom scale is a standardized assessment tool consisting of 22 questions ascertaining the severity of symptoms associated with a concussion. Players were led through the questionnaire by an investigator experienced in assessing patients in this manner. Responses for each symptom were rated by the subject on a seven-point Likert scale (zero, no symptom; and 6, severe symptom).
Head impact exposure
Head impact exposure was measured during every practice and game using the HIT system and the Sideline Response System (SRS) (Riddell Corp, Elyria, OH). The HIT system consists of an encoder unit with six spring-mounted single-axis accelerometers, an onboard data acquisition (8 bit; 1000 Hz per channel) and memory storage device (up to 100 impacts), and a wireless transceiver (903–927 MHz). The HIT encoders were fitted within Riddell Revolution Speed helmets. When one accelerometer registered an impact equal to or greater than a predefined threshold, 40 ms of data were transmitted from all six accelerometers to a sideline unit and estimations of resultant linear acceleration, rotational acceleration, and impact location were generated by the HIT system using a novel algorithm (7). Only impacts with peak resultant linear accelerations of 10g or more were included in the data analysis. Simple descriptive analyses of HIE data were performed to generate summaries of impact frequency, magnitude, and location. Cumulative indices of HIE for each player were calculated as impact frequency (#) × impact magnitude (linear acceleration [g force]). Cumulative values were calculated for the entire season, the last week of the season, and the game of the POST-G evaluation.
Associations between individual HIE and differences in PRE to POST and PRE to POST-G were analyzed using covariate-adjusted differences in SAS Version 9.3 (SAS Institute, 2009). Initially, simple group differences and appropriate statistical tests were examined. Head impacts were subsequently added to the model as covariates for group differences. Both the test of the difference and the covariate-adjusted test were examined in PROC GLM. Each variable of interest was examined separately. All data are expressed as mean ± SD. The level of significance for all comparisons was set as α < 0.05.
Head impact exposure
There were a total of 6183 head impacts measured among all participants for the entire season, with 3787 impacts (61%) occurring in 27 practices, 2037 impacts (33%) occurring in nine games, and 359 impacts (6%) occurring in nine pregame warm-ups. The median (maximum) head impact counts for the entire team were 136 (317) impacts in practices and 246 (297) impacts in games, with considerable variability between sessions. The median head impact frequency per player for the entire season was 252 impacts when including all participants, 269 impacts for players who participated in 33% or more of all sessions, and 306 impacts for those who participated in 66% or more of all sessions. For the entire season, the median head impacts per player per session were nine impacts per practice and 12 impacts per game. The median number of head impacts varied greatly from session to session, with a median range of 3–22 impacts per practice and 6–18 impacts per game (Fig. 1). The maximum head impacts experienced by a single player in one session were 50 (game) and 54 (practice), whereas the maximum number recorded for a single player over the entire season was 880 impacts.
Linear acceleration for all registered head impacts ranged from 10.0g to 175.9g, with a mean of 25.5g and a median of 20.2g. For practices, the linear head impact acceleration range was 10.0g–175.9g, whereas the mean and median were 25.0g and 19.9g, respectively. For games, the linear head impact acceleration range was 10.0g–154.1g, whereas the mean and median were 26.8g and 20.9g, respectively. The 95th percentile for linear head impact acceleration was 57.3g in all sessions, 55.0g in practices, and 63.0g in games. Overall distribution frequencies for linear head impact acceleration are shown in Figure 2. Rotational acceleration for all registered head impacts ranged from 7.1 to 12,322.5 rad·s−2, with a mean of 1691.8 rad·s−2 and a median of 1407.4 rad·s−2. For practices, the rotational head impact acceleration range was 7.1–12,322.5 rad·s−2, whereas the mean and median were 1628.6 and 1383.0 rad s−2, respectively. For games, the rotational head impact acceleration range was 9.5–10,795.7 rad·s−2, whereas the mean and median were 1832.8 and 1448.5 rad·s−2, respectively. The 95th percentile for rotational head impact acceleration was 3929.0 rad·s−2 for all sessions combined, 3709.6 rad·s−2 in practices, and 4782.3 rad·s−2 in games. Overall distribution frequencies for rotational head impact acceleration are shown in Figure 2.
Head impacts occurred most frequently to the front of the helmet (42.5%) followed by the back of the helmet (26.1%), the side of the helmet (18.1%), and the top of the helmet (13.3%) (Fig. 3). The frequency distribution was nearly identical for practices and games. By impact location, the highest median linear acceleration (22.6g) and the highest 95th percentile acceleration (70.5g) were measured from impacts to the top of the helmet. Impacts to the front of the helmet had the highest median rotational acceleration (1615.0 rad·s−2), whereas impacts to the back of the helmet had the highest 95th percentile rotational acceleration (5084.4 rad·s−2).
There were no PRE versus POST or PRE versus POST-G impairments among any measures of neurocognitive function, balance, oculomotor performance, or SRT. In addition, no statistically significant associations were discovered between individual HIE during one game, the last week of the season, or the entire season and deficits in any of the neurologic measures.
One 12-yr-old player (CP) experienced a concussion during the seventh game of the season. CP received separate head impacts from two opposing defenders while carrying the ball on offense. One head impact was located superior to the left eye and the other superior and lateral to the right eye, both in the forehead region. The HIT system timestamp indicated the head impacts occurred 0.54 s apart and had linear accelerations of 52.2g and 52.4g and rotational accelerations of 4508 and 4646 rad·s−2, respectively (Table 1). Before that game, CP had accumulated 242 head impacts during the season, but these were the first two head impacts he experienced that day. A summary of the player’s HIE data for the entire season is shown in Table 2.
CP did not suffer loss of consciousness but remained on the ground for a short period of time and was assisted off the field under his own power. He later reported dizziness, fogginess, having a headache, and feeling slowed down immediately after the injury and did not return to the game. CP was evaluated approximately 1 h post-injury (PI-0) in conjunction with our planned POST-G assessment. Upon examination, CP was alert and oriented but seemed dazed and answered questions slowly. He had no focal neurologic deficits or amnesia; extraocular movements were intact; and pupils were equal, round, and reactive to light; and his immediate memory and short-term recall were normal. CP returned to our clinic on postinjury days 2 (PI-2), 6 (PI-6), 12 (PI-12), 20 (PI-20), and 27 (PI-27) before he was cleared for return to full activity.
CP’s KD-T time at PI-0 was markedly slower than his PRE performance (89.59 vs 50.75 s), but it progressively improved at subsequent visits and returned to his PRE value by PI-20 (Fig. 4). Simple RT at PI-0 was nearly twice as slow as PRE (0.384 vs 0.216 s) but paralleled the improvements on the KD test and also reached his PRE value by PI-20 (Fig. 4). CP’s postural control was assessed with the WBB at PI-0 and by WBB, FP, and NC at all subsequent PI visits (Fig. 4). At PI-0, COP excursion on WBB was moderately greater than at PRE but continued to increase until reaching a peak at PI-6; it then returned to his PRE value by PI-20. This pattern of balance recovery was identical when assessed by FP; however, postural stability assessed by NC remained similarly impaired on PI-2 through P-12, with no nadir, before returning to his PRE values by PI-20. All ImPACT composite scores demonstrated severe impairment at PI-2 but had returned to PRE by either PI-12 (VMS) or PI-20 (VeM, ViM, and RT). CP reported multiple postconcussive symptoms on the PCSS, which gradually dissipated by PI-27.
We found that head impact magnitude and location profiles among 11- to 13-yr-old football players were similar to what has been reported in youth, high school, and collegiate football over one season, although the players in this study had intermediate impact frequencies than what has been reported for younger and older players. This is consistent with our hypothesis that HIE in adolescent football players would be lower compared to high school and collegiate players. It is notable, however, that impact magnitude characteristics were nearly identical to those of high school players in previous studies. In contrast to our second hypothesis, there were no statistically significant associations between individual HIE during a single football season and short-term changes in select clinical measures of neurologic function that would indicate impairment. Finally, one subject experienced a concussion after receiving two head impacts in close temporal and spatial proximity that were in the moderate range of linear acceleration registered throughout the season. This occurrence brings attention to the risk of improper and potentially misleading on-field injury detection when relying on “concussion thresholds” for head impact monitoring that have not been consistently established or validated.
There seems to be a progressive increase in HIE relative to age among youth football players. This pattern was revealed for both head impact frequency and magnitude. In a 2-yr study of 7- to 8-yr-old football players, the mean impact frequency during a season was 161 head impacts per player (36); whereas among players 9–12 yr old, a mean of 240 head impacts was reported (6). Among our 11- to 13-yr-old players, we observed a mean of 252 head impacts per player per season; and in a group of slightly older 12- to 14-yr-old players, a mean impact frequency of 275 impacts per player was measured (11). The mean impacts per session were similar in all studies (6–10 impacts per practice and 11–12 impacts per game), indicating that the primary reason for higher head impact frequencies during a football season with increased age is that older players typically participate in more scheduled sessions (practices and/or games).
Among high school and collegiate players, head impact frequency is more variable owing largely to positional differences. Broglio et al. (3) reported a mean head impact frequency of 652 impacts per player per season in a multiyear study of high school football players. The highest individual head impact frequency per season in that study was 2235 impacts. In a study of high school and collegiate football players, Schnebel et al. (31) reported a mean seasonal head impact frequency of 520 impacts per player at the high school level and 1354 impacts per player at the collegiate level. In larger samples of collegiate football players, Guskiewicz et al. (20) indicated that individual head impact frequency was approximately 950 impacts per season, whereas Crisco et al. (9) reported a median seasonal head impact frequency of 420 impacts per player. The latter group also reported that one player in their study received 2492 impacts in a single season. Despite the variability in these values, these studies indicate that high school and collegiate football players consistently experience a higher head impact frequency than younger players. On a per-session basis, several authors have also reported higher head impact rates among high school and collegiate football players than what our data reveal. Among high school players, the difference seems to be more a function of higher head impact frequencies in games, as Broglio et al. (4) and Urban et al. (34) found 9.2 and 9.4 head impacts per practice, and 24.5 and 15.5 impacts per game, respectively. The individual head impact frequency in collegiate games has also been shown to be higher (14.3 impacts per game) than the head impact rates of our youth players (8). Still, some players in our study had head impact frequencies in games that matched or exceeded the mean values of these older groups.
Consistent with similar investigations, head impact magnitudes were characterized by a skewed frequency distribution, with approximately 50% of head impacts having linear accelerations between 10g and 20g. Still, there were 103 impacts that registered 80g or more, which is a magnitude that has been previously defined as “high” (25). This suggests that whereas most head impacts in youth football are of a relatively low or moderate force, there are many impacts over the course of a season that would be considered high magnitude at any level of play.
Previous studies of youth football players have shown that the median linear accelerations were lower compared to reported values for high school players. In a study of 7- to 8-yr-old players, Young et al. (36) found a median head impact of 16g. Among 9- to 12-yr-old players, Cobb et al. (6) reported a median head impact of 18g. By comparison, in one of the largest high school studies of HIE, Eckner et al. (15) reported a median head impact of 21g. The median head impact of the players in our study (11–13 yr) was 20g, indicating that there is a progressive increase in head impact force with age among youth and high school players. Recently, however, Daniel et al. (11) measured a median head impact of 22g among 12- to 14-yr-old players. It is not clear if the slightly higher impact magnitude of these middle school players is clinically relevant or simply reflects similarity within a narrow range of head impact forces experienced by middle school and high school players.
Larger differences between high school and youth players have been shown in the frequency of higher magnitude impacts. The 95th percentile impact forces for 7- to 8-yr-old, 9- to 12-yr-old, 12- to 14-yr-old, and high school players were 38g, 43g, 60g, and 56g, respectively (6,11,15,36). The 11- to 13-yr-old players in our study had a 95th percentile impact force of 57g. In games, this value was 63g, which matches the 95th percentile impact force that was reported for collegiate players (9). The median rotational acceleration of head impacts in the current study (1407 rad·s−2) was considerably greater than what was reported for 7- to 8-yr-old (621 rad·s−2), 9- to 12-yr-old (856 rad·s−2), and 12- to 14-yr-old (987 rad·s−2) players (6,11,36). However, both median and 95th percentile (3929 rad·s−2) rotational accelerations were nearly identical to that of high school players (1394 and 3901 rad·s−2, respectively) (15). Collectively, these data show that the head impact forces of the players in this study, on a hit-by-hit basis, were nearly identical to those of high school players.
There is remarkable consistency in the location of head impacts among players in this and other youth football studies (6,36). The highest impact frequencies occur to the front of the helmet followed by the back, side, and top of the helmet. The frequency distributions of impact location are also similar to what has been reported in the high school ranks (34). These observations suggest that head impact location may largely be independent of playing age and/or experience, or there may be insufficient instruction of blocking and tackling techniques that positively affect head impact location as youth players develop.
We have shown previously that youth football players who were the same age as the participants in this study did not exhibit clinically recognized signs of neurologic impairment after a single season of play (28). Whereas this was the first study to examine neurologic function in the context of youth football participation, we did not perform biomechanical assessments of HIE, so we were unable to probe for individual associations between HIE and changes in test performance. Consistent with our previous investigation, no neurologic deficits were detected among the clinical measures used in the current study. Furthermore, there was no identifiable link between HIE, when calculated for either the entire season or the last week of the season, and changes in individual assessment scores from PRE to POST. A subset of tests was performed after one game with PRE to POST-G scores analyzed in relation to both HIE in the game preceding the evaluation and HIE up to that point in the season. Similar to the findings for the entire season, HIE did not influence POST-G scores in oculomotor performance, SRT, or balance.
Whereas Miller et al. (26) similarly reported no decreases in neurocognitive scores in collegiate football players when comparing preseason, midseason, and postseason values, other studies have shown impairments in neurocognitive function and/or balance among nonconcussed high school and collegiate football players, including an association between HIE and cognitive deficits (2,27,32). Furthermore, data indicating neurophysiological changes in the brains of nonconcussed football players are beginning to emerge (2,12,32). Considering that cumulative HIE over a season was less in our youth players than in high school football, players in this study and in our previous work may have remained below some yet unidentified threshold for readily identifiable clinical impairment. Although football does seem to have a detrimental effect on the short-term neurologic function of some, but not all, nonconcussed players, the prevalence of this condition and factors related to individual susceptibility (or resilience) require further study. It is also unknown if acute neurologic deficits in nonconcussed football players are associated with neuropathology later in life.
It is possible that the measures of neurologic function we used were neither sensitive nor comprehensive enough to detect impairments. However, the assessment protocol we used meets or exceeds the standards that are recommended for the clinical evaluation of concussion (22). Another explanation that cannot be ruled out is the existence of a transient neurologic deficit during the season that recovered by the POST assessment. Moreover, the absence of functional impairments in this younger player population compared to what has been found in older players could be the result of greater neuroplasticity that allows for a more favorable adaptive response to repetitive impact-related brain trauma. Further investigation is needed to determine if an age-specific recovery process occurs.
There was one physician-diagnosed concussion in this study that resulted from two head impacts of close spatial and temporal proximity during a game. Both impacts occurred to the superior frontal area of CP’s head, approximately 0.5 s apart. Head impacts to the frontal area of the helmet accounted for more concussions than any other location among more than 1000 high school and collegiate football players wearing helmets instrumented with the HIT system in previous studies (1). However, this is likely a function of the frontal region of the helmet having a greater impact frequency rather than a susceptibility to brain injury from impacts to that region.
The linear accelerations of both concussion-related impacts were remarkably similar at 52.2g and 52.4g, respectively. These impacts were considerably lower than the peak linear acceleration (season) of CP (113.6g) and of all players on the team (175.9g) but were relatively high as a percentile of all head impacts during the season (>93rd percentile for both CP and the entire team). Whereas the magnitude of both concussion-related impacts were well below the average peak linear acceleration (102.5g) of head impacts that preceded concussion incidents among dozens of high school and collegiate players, they both fit within the range of linear accelerations that was reported (29.3g–205.3g) (1). The concussion incurred by CP in our study is consistent with other reported concussions in the literature that have been associated with linear accelerations well below 80g, which has been defined as high magnitude (10,25). Thus, people should not rely upon commercial impact sensors to alert them of an on-field concussion based on a particular impact threshold being registered, as a validated concussion threshold does not exist and concussions can surely occur at impact forces lower than what may be perceived as “high.”
Of note, both of CP’s concussion-related rotational accelerations (4508 and 4656 rad·s−2) were greater than the peak rotational accelerations of the concussed high school and collegiate athletes (3977 rad·s−2) of Beckwith et al. and were above the 96th and 97th percentiles of all impacts CP and his entire team experienced, respectively. Whereas these data suggest that either impact alone may have been sufficient to cause a concussion, the potential additive effect of two similarly high-magnitude impacts in such close temporal and spatial proximity may also have measurably contributed to this injury. To our knowledge, a concussive dual-impact biomechanics profile of this nature has not previously been reported.
All objective testing measures for CP were impaired immediately after his concussion and did not return to baseline until PI-20. However, PCSS was elevated above baseline until PI-27; thus, all objective markers returned to normal before complete resolution of clinical symptoms. This time course of recovery was considerably longer than a mean symptom resolution of 5.9 d after concussion in 89 high school and collegiate football players reported by Beckwith et al. (1) although less than the maximum period of symptom resolution (59 d) reported in this study. A protracted recovery from concussions among children and adolescents compared to adults, consistent with this case study observation, has previously been identified in the literature (33).
Limiting unnecessary contact in practices has been suggested as a means to reduce HIE in football (6). Presumably, less contact is likely to reduce the overall burden of head impacts, and this is much easier to achieve in a controlled practice setting than in a competitive game situation. Cobb et al. (6) measured HIE in three youth football teams and found that impact frequency was substantially reduced on a team that intentionally reduced contact time in practices, consistent with a new rule issued by a major youth football organization. Players on the contact-restricted team had fewer head impacts per practice (6.2 impacts) than players on either of the other teams (12.9 and 9.5 impacts, respectively). Whereas reducing impact frequency seems to be a logical and reasonable approach, attempts to reduce head impact forces through the instruction and use of proper blocking and tackling techniques is also recommended, although the effectiveness of such attempts is less certain at this time.
This study has several limitations. First, the findings are limited to a relatively small sample of similar-age youth football players from a single team. Thus, these HIE data cannot be generalized to all of youth football or even this specific age group. It is likely that tremendous variation in HIE exists among youth football participants owing to differences in age and maturation, coaching style and philosophy, league rules, game and practice schedules, and individual ability and motivation.
Preseason, POST, and POST-G assessments were limited to practical, clinical measures of concussion that may lack the precision and sensitivity to detect subtle changes in neurologic function. Investigating neurophysiological alterations in the brain through the use of sophisticated imaging techniques may have aided our examination of potential injury but was beyond the scope of this study. Low subject participation at the POST evaluation further limited the statistical power to probe for neurologic deficits and potential associations with HIE. Although we did not discover adverse changes in any selected measure of neurologic function among our study population, the use of an age-matched control group of individuals not participating in a contact-collision sport would have provided an opportunity to verify that these results were normal.
This is the first study to investigate HIE and clinical measures of neurologic function in football players younger than high school age. There were no apparent associations between one season of HIE and short-term changes in clinical measures of neurologic function that would indicate neurocognitive or neuromotor impairment. Although the 11- to 13-yr-old players in this study had a lower head impact frequency than what has previously been reported for high school players, on a hit-by-hit basis, these younger players experienced similar head impact forces as their older counterparts. One player experienced a concussion after receiving two head impacts in rapid succession that were in the moderate range of magnitudes registered throughout the season. This injury brings attention to the risk of relying on concussion thresholds provided by impact monitoring devices that have not been validated. Future research is needed to generate comprehensive HIE data for youth football that can be used to create and evaluate evidence-based recommendations for reducing unnecessary impact exposure and minimizing the risk of brain injury in this population. Additional efforts aimed at determining injury thresholds and exploring long-term consequences of repetitive head impacts over multiple seasons are warranted.
Sanford Research, Sioux Falls, SD, provided financial support for this work.
The authors thank Shanna Kindt, MS, Hannah Nelson, MS, Tryg Odney, MS, and Caitlin Pearl, MS, for their assistance with data collection. The authors also sincerely thank the research volunteers and their parents for their participation in this study, and they greatly appreciate the support of this project by South Dakota Junior Football, Inc.
The authors have no conflicts of interest to declare.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Beckwith JG, Greenwald RM, Chu JJ, et al. Timing of concussion
diagnosis is related to head impact exposure prior to injury. Med Sci Sports Exerc
. 2013; 45 (4): 747–54.
2. Breedlove EL, Robinson M, Talavage TM, et al. Biomechanical correlates of symptomatic and asymptomatic neurophysiological impairment in high school football. J Biomech
. 2012; 45 (7): 1265–72.
3. Broglio SP, Eckner JT, Martini D, Sosnoff JJ, Kutcher JS, Randolph C. Cumulative head impact burden in high school football. J Neurotrauma
. 2011; 28 (10): 2069–78.
4. Broglio SP, Sosnoff JJ, Shin S, He X, Alcaraz C, Zimmerman J. Head impacts during high school football: a biomechanical assessment. J Athl Train
. 2009; 44 (4): 342–9.
5. Centers for Disease Control and Prevention. Nonfatal traumatic brain injuries related to sports and recreation activities among persons aged ≤19 years—United States, 2001–2009. MMWR Morb Mortal Wkly Rep
. 2011; 60 (39): 1337–42.
6. Cobb BR, Urban JE, Davenport EM, et al. Head impact exposure in youth football: elementary school ages 9–12 years and the effect of practice structure. Ann Biomed Eng
. 2013; 41 (12): 2463–73.
7. Crisco JJ, Chu JJ, Greenwald RM. An algorithm for estimating acceleration magnitude and impact location using multiple nonorthogonal single-axis accelerometers. J Biomech Eng
. 2004; 126 (6): 849–54.
8. Crisco JJ, Fiore R, Beckwith JG, et al. Frequency and location of head impact exposures in individual collegiate football players. J Athl Train
. 2010; 45 (6): 549–59.
9. Crisco JJ, Wilcox BJ, Beckwith JG, et al. Head impact exposure in collegiate football players. J Biomech
. 2011; 44 (15): 2673–8.
10. Daniel RW, Rowson S, Duma SM. Head impact exposure in youth football. Ann Biomed Eng
. 2012; 40 (4): 976–81.
11. Daniel RW, Rowson S, Duma SM. Head impact exposure in youth football: middle school ages 12–14 years. J Biomech Eng
. 2014; 136 (9): 094501.
12. Davenport EM, Whitlow CT, Urban JE, et al. Abnormal white matter integrity related to head impact exposure in a season of high school varsity football. J Neurotrauma
. 2014; 31 (19): 1617–24.
13. Dorman JC, Valentine VD, Munce TA, Tjarks BJ, Thompson PA, Bergeron MF. Tracking postural stability of young concussion
patients using dual-task interference. J Sci Med Sport
. 2015; 18 (1): 2–7.
14. Duma SM, Rowson S. Past, present, and future of head injury research. Exerc Sport Sci Rev
. 2011; 39 (1): 2–3.
15. Eckner JT, Sabin M, Kutcher JS, Broglio SP. No evidence for a cumulative impact effect on concussion
injury threshold. J Neurotrauma
. 2011; 28 (10): 2079–90.
16. Elbin RJ, Schatz P, Covassin T. One-year test-retest reliability of the online version of ImPACT in high school athletes. Am J Sports Med
. 2011; 39 (11): 2319–24.
17. Galetta KM, Barrett J, Allen M, et al. The King-Devick test as a determinant of head trauma and concussion
in boxers and MMA fighters. Neurology
. 2011; 76 (17): 1456–62.
18. Galetta KM, Brandes LE, Maki K, et al. The King-Devick test and sports-related concussion
: study of a rapid visual screening tool in a collegiate cohort. J Neurol Sci
. 2011; 309 (1–2): 34–9.
19. Gessel LM, Fields SK, Collins CL, Dick RW, Comstock RD. Concussions among United States high school and collegiate athletes. J Athl Train
. 2007; 42 (4): 495–503.
20. Guskiewicz KM, Mihalik JP. Biomechanics
of sport concussion
: quest for the elusive injury threshold. Exerc Sport Sci Rev
. 2011; 39 (1): 4–11.
21. Guskiewicz KM, Valovich McLeod TC. Pediatric
. 2011; 3 (4): 353–64.
22. Harmon KG, Drezner JA, Gammons M, et al. American Medical Society for Sports Medicine position statement: concussion
in sport. Br J Sports Med
. 2013; 47 (1): 15–26.
23. Lovell MR, Iverson GL, Collins MW, et al. Measurement of symptoms following sports-related concussion
: reliability and normative data for the post-concussion
scale. Appl Neuropsychol
. 2006; 13 (3): 166–74.
24. Meehan WP 3rd, d’Hemecourt P, Collins CL, Taylor AM, Comstock RD. Computerized neurocognitive testing for the management of sport-related concussions. Pediatrics
. 2012; 129 (1): 38–44.
25. Mihalik JP, Bell DR, Marshall SW, Guskiewicz KM. Measurement of head impacts in collegiate football players: an investigation of positional and event-type differences. Neurosurgery
. 2007; 61 (6): 1229–35.
26. Miller JR, Adamson GJ, Pink MM, Sweet JC. Comparison of preseason, midseason, and postseason neurocognitive scores in uninjured collegiate football players. Am J Sports Med
. 2007; 35 (8): 1284–8.
27. Mulligan I, Boland M, Payette J. Prevalence of neurocognitive and balance deficits in collegiate aged football players without clinically diagnosed concussion
. J Orthop Sports Phys Ther
. 2012; 42 (7): 625–32.
28. Munce TA, Dorman JC, Odney TO, Thompson PA, Valentine VD, Bergeron MF. Effects of youth football on selected clinical measures of neurologic function: a pilot study. J Child Neurol
. 2014; 29 (12): 1601–7.
29. Oride MK, Marutani JK, Rouse MW, DeLand PN. Reliability study of the Pierce and King-Devick saccade tests. Am J Optom Physiol Opt
. 1986; 63 (6): 419–24.
30. Schatz P, Pardini JE, Lovell MR, Collins MW, Podell K. Sensitivity and specificity of the ImPACT Test Battery for concussion
in athletes. Arch Clin Neuropsychol
. 2006; 21 (1): 91–9.
31. Schnebel B, Gwin JT, Anderson S, Gatlin R. In vivo study of head impacts in football: a comparison of National Collegiate Athletic Association Division I versus high school impacts. Neurosurgery
. 2007; 60 (3): 490–5.
32. Talavage TM, Nauman EA, Breedlove EL, et al. Functionally detected cognitive impairment in high school football players without clinically diagnosed concussion
. J Neurotrauma
. 2014; 31 (4): 327–38.
33. Toledo E, Lebel A, Becerra L, et al. The young brain and concussion
: imaging as a biomarker for diagnosis and prognosis. Neurosci Biobehav Rev
. 2012; 36 (6): 1510–31.
34. Urban JE, Davenport EM, Golman AJ, et al. Head impact exposure in youth football: high school ages 14 to 18 years and cumulative impact analysis. Ann Biomed Eng
. 2013; 41 (12): 2474–87.
35. Valentine VD, Dorman JC, Munce TA, et al. Balance testing of concussion
patients using dual-task interference. Med Sci Sports Exerc
. 2012; 44 (5 Suppl): S264–5.
36. Young TJ, Daniel RW, Rowson S, Duma SM. Head impact exposure in youth football: elementary school ages 7–8 years and the effect of returning players. Clin J Sport Med
. 2014; 24 (5): 416–21.
Keywords:© 2015 American College of Sports Medicine
BIOMECHANICS; CONCUSSION; INJURY RISK; PEDIATRIC