High school and youth

According to FIFA, there are 265 million registered soccer players worldwide, including 26 million women, and 22 million registered youth soccer participants,5 with 4.2 million coming from the USA (2.5 million male, 1.7 million female). Between 2000 and 2006, there was a 54% and 21% increase worldwide in the number of female and male participants registered, respectively.5 Of the youth players in the USA, 19 721 players participate in Developmental Academies (DA),6 which includes several age groups ranging from under (U)-12 to U-18–19 years, for both male and female players, participating in elite player development as part of US Soccer. In the 2017–2018 season, the DA programmes included 896 teams in 197 clubs, with over 13 000 games.6

Faude et al summarised the literature regarding injuries to children and adolescent soccer players (aged 5–19 years) and reported that about 5% of all injuries in these age groups were to the head.7 The percentage of concussions ranged between 1% and 7% across studies with the majority reporting 2% or fewer. Faude et al followed up their review with a prospective project of children aged 7–12 years focusing their attention on 39 head injuries (5% of all 791 injuries), of which 11 (28% of all head and neck injuries) were classified as concussion.8 Seventy-five per cent of injuries were due to contact with another player while ball contact was cited as the injury mechanism for two concussions. In addition, the authors noted that the rate of concussions (per 1000 hours of soccer exposure) increased slightly with age (U-9=0.04; U-11=0.07; U-13=0.06).

In a more recent online survey of 8104 youth soccer teams including 101 699 players (59 621 boys; 42 048 girls) aged 7–14 years, (Kontos, 2017, personal communication), the overall reported concussion incidence rate was 0.85/1000 AE. In this study, concussions were 5.7 times more likely to occur in games (1.73/1000 AEs) than practices (0.27/1000 AEs). Although there were no significant differences in concussions between boys and girls, older (aged 11–14 years) players (1.08/1000 AEs) were 2.7 times more likely to have concussions than younger (aged 7–10 years) players (0.40/1000 AEs). Only one in five concussions were from attempted soccer heading and no concussions from heading were reported in players younger than 12 years of age that may reflect adherence to recent US Soccer mandated restrictions on heading in young players. Representative data for youth through high school-aged soccer players are presented in table 1.

The overall rate of SRC in high school soccer in the USA from 2005/2006 to 2013/2014 was estimated to be 0.36/1000 AEs (girls=0.45/1000 AEs; boys=0.28/1000 AEs) with a higher rate in games than in practices in girls (rate ratio=15.1) compared with boys (rate ratio=11.0).9 The overall rates for high school girls soccer trail American football (0.64/1000 AEs), boys ice hockey (0.54/1000 AEs) and boys lacrosse (0.40/1000 AEs).10 During the 2014–2015 academic year, concussions represented the highest proportion of all injuries in girls high school sports.11

Among youth and high school players, concussion is the second most common match injury representing approximately 24% of all injuries.12 While concussion rates increased each year, there was a slight increase beginning in 2006,13 which continued to rise yearly14 until plateauing in 2011/2012.9 These trends likely reflect greater awareness, better recognition and increased reporting of concussion.

In a retrospective study of high school and youth soccer players, nearly 60% of concussions involved contact or collision with another player.9 The player behaviour of ‘heading the ball’ represented 28% of all concussions in high school soccer, 70% of which were from player-to-player contact, not ball-to-player contact.9 Of the concussions that reflected ball-to-head mechanisms, it is unclear just how many involved successful purposeful heading of the ball; contact may have been purposeful or accidental. Previous prospective reports in collegiate players found that no concussions were due to intentional, purposeful heading.15 Recently, researchers reported that 75% of concussions in youth and high school soccer players occurred when the player was unaware of the oncoming contact or collision.16 Concussions from unexpected collisions and contact led to longer symptom duration than when a player was aware of (and presumably braced for) the oncoming contact or collision. These findings suggest situational awareness may influence concussion risk and clinical outcomes following injury.

In spite of improvements in concussion awareness and recognition, as many as 44% of youth soccer players indicated that they would not report their concussion symptoms.17 This is particularly concerning as researchers recently reported that youth sport participants aged 12–18 years who continued to play after a concussion were nearly nine times more likely to have a prolonged (>21 days) recovery.18 Taubman et al 19 reported similar findings, with a greater likelihood of prolonged recovery (>30 days) in youth aged 11–19 years who did not rest immediately after concussion versus those who did rest (67% vs 35%). On a positive note, in a recent independent survey, 92% of players were reported to have sought clinical care for their concussion (Kontos, 2017, personal communication).

Collegiate

One of the earliest surveillance studies on concussions in college age, male and female players reported the concussion rate to be 0.6/1000 AE for men and 0.4/1000 AE for women. The two most common mechanisms of injury, accounting for half of all concussions, were head-to-head and accidental ball-to-head impacts. No concussions were due to routine, purposeful, successful heading of the ball.15 Rates of concussion have risen as injury definitions, injury recognition and reporting have evolved.

The National Collegiate Athletics Association (NCAA) has been the primary source of injury data for collegiate level players since 1986.20 Their data indicate that concussion rates are higher in competition when compared with practice in both women’s (rate ratio: 8.4–9) and men’s soccer (rate ratio: 5.5).21 22 In comparison with other sports, the NCAA data show the competition concussion rate for men’s soccer is less than wrestling, American football and ice hockey while women’s soccer trails only ice hockey.22 When comparing the match injury rates for women and men, women had a higher rate for both head and face injuries (3.06/1000 AE vs 2.19/1000 AE) as well as concussion (1.83/1000 AE vs 0.94/1000 AE).21 Representative data for collegiate soccer players are presented in table 2.

Injury mechanisms are also available from the NCAA Injury Surveillance Programme. Summaries of 11 years of data for both males and females depict that the majority of concussions do not occur during successful, purposeful heading. In those cases where concussions are associated with heading, the vast majority are due to player-to-player contact rather than player contact with the ball.21 22 The proportion of heading-related concussions due to contact with the ball is larger in women’s soccer than in men’s soccer (27.4% vs 10.9%; Parsons, 2017, personal communication). Unfortunately, the NCAA data do not identify accidental ball-to-head impacts.

Professional

Data that are based on professional players come from two similar, but unique settings. The first is the domestic professional season (eg, the English Premier League, La Liga, MLS, etc). The second setting is a tournament (FIFA World Cup (U-17 through full national teams; male and female), Olympics, UEFA Champions League, EURO, etc). While both settings are highly competitive, tournaments are generally assumed to represent the highest and most intense level of competition for the respective age and gender. Data from domestic leagues and tournaments will be presented separately.

Recognise and remove, assessment, treatment and recovery

The recognition, assessment and treatment of SRC in soccer is important, and for the purposes of the HIS, the findings from the fifth International CISG Consensus Statement are supported and the reader is referred to the consensus statement1 as well as the accompanying systematic reviews for a full discussion of these issues. The focus of this section is to review areas that might present unique sport-specific applications or considerations for soccer.

Case study: the rugby experience

Rugby is a sport similar to soccer in that it is fast-paced with a similar number of players participating on a large pitch, with ongoing play with limited stoppage time. Rugby is different than soccer in that it is a tackling sport and an interchange sport (ie, temporary player substitutions permitted), while soccer is a replacement sport (ie, temporary substitutions not permitted).

In 2012, World Rugby instituted a rule to allow temporary substitution, a mandatory and maximum of 10 min, for elite players with suspected SRC. A ‘Head Injury Assessment’ (HIA) was developed that contained a three-point process.25

1. Level one (HIA 1) is an immediate postinjury multimodal evaluation that does not diagnose concussion, but uses the following four components to identify suspected concussion: a) criteria that warrant immediate and permanent removal from play (convulsion, tonic posturing, ataxia, suspected or confirmed LOC, clearly dazed, oculomotor signs, disorientation, confusion, any behaviour change or any sign or symptom of concussion when the athlete is evaluated), b) an off-field screen, c) video review, d) clinical opinion.

2. HIA 2 is completed within 3 hours of injury and consists of the SCAT5 and clinical interview.

3. The HIA 3 is completed 36–48 hours postinjury and consists of SCAT5 symptoms, balance, cognitive assessment (SAC at a minimum or a computerised neurocognitive test).

Utilisation of HIA has improved recognition and management of injury at the elite level of play. The addition of the HIA2 and HIA3 components incorporates serial assessments and decreases the likelihood of missing injuries associated with delayed symptoms.26 58 59 The rugby experience at the elite level is a helpful guide for soccer and likely illustrative of the possibilities for elite play. Similar to soccer, the emphasis for community play in rugby has been on ‘recognise and remove’ principles. Emphasis has also been placed at all levels to remove head-to-head contact with a ‘zero tolerance’ policy and lower level of tackle, avoiding head contact.

Biomechanics of heading

An understanding of heading begins with tracking the flight of the ball, which leads to multiple decisions that experienced players must make very quickly. Youth players often struggle with predicting the flight of the ball and then having to coordinate subsequent body actions and movements necessary to intercept the ball, all of which must happen before any head-ball contact. This entry level skill, tracking the ball, is something coaches can teach with the understanding that the ability to track a ball in flight (and the myriad of accompanying movement responses) takes time to fully develop and coordinate. When considering child development, research would support that heading is a complex motor skill that most children might not learn, much less master, until age 10–12 years.60–64 This consideration of developmental trajectories is consistent with guidance issued by US Soccer regarding age-related heading restrictions (see page 11 of online supplementary appendix C).49

The skill of heading has been described as consisting of three stages: preparation, contact and follow through65 and has been studied in laboratory settings, primarily to measure impact forces associated with heading the ball. The forces measured during the act of heading at the youth level are between 4.5 and 62.9 g,65 66 approximately 37.9 g in games at the high school level67 and under 43 g in collegiate soccer studies.68–70 While these data are useful in quantifying impact forces, there are concerns related to the sensitivity and specificity of sensor data,70 as well as the correlation, if any, to concussive injury. Nonetheless, by most calculations, forces associated with successful purposeful heading seem to be well below most thresholds of head injury.50 Applying these measurements during match play is a work in progress.

Frequency of heading

Youth

Data regarding heading in the youth game are limited. What should be a fairly simple task (counting head-ball impacts) can be laborious. Numerous methods have been used such as live or video observers, extrapolating a short segment to the full match and player or coach self-report. Self-report of head-ball impacts has recently been shown to overestimate observer counts by an average of 51%.71 Among youth players aged 13 years and under heading is rare; when it does occur, it is often the result of balls that are thrown in, bounced from the ground or unintentional deflections.

In a group of male and female players aged 9–15 years, it was reported that there was an average of 1.64 headers per game.72 A trend for an increase in the number of headers per game as players got older was reported, where the rates for headers per game were 0 for children who are aged 9 years, 0.65 for 10 years, 1.56 for 11 years, 1.11 for 12 years, 2.65 for 13 years and 2.16 for 14 years.72 Later studies involving female youth players report heading frequencies ranging from 0.93 headers per game73 to 1.95 headers per game.74

A recent retrospective online survey study of youth soccer players (from U-8 to U-15 age groups, boys and girls) asked representatives of 8104 teams from all 55 US Soccer state associations during the fall of 2016 season to determine the occurrence of headers at the youth level. Team representatives who responded to the survey included coaches, team managers and designated adult team contacts (Kontos, 2017, personal communication). This study revealed that approximately 50% of games at the youth level do not have any headers, approximately 94% of all games have 1–10 headers and <1% of all games have between 21 and 30 headers (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Percentage of survey respondents with regard to the question as to the number of headers per week by teams in practice in given range categories including none (blue), 1–10 (red), 11–20 (orange), 21+ combined (grey). This included 0.76% (21–30 range) and 0.15% (31+ range).

The frequency of self-reported heading during youth practices is similar to those seen during games. Approximately 60% of practices do not include any headers. In those practices where heading was reported it is likely that the practices incorporated heading-specific skill training (figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Percentage of survey respondents with regard to the question as to the number of headers per week by teams in practice in given range categories including none (blue), 1–5 (red), 6–10 (orange), 11+ combined (grey). This included 3.35% (11–15 range), 1.74% (16–20 range) and 1.10% (21+ range).

Individual player self-report has also been used to estimate the frequency of heading impacts in youth, collegiate and amateur players.75–77 Player self-report has been (very recently) shown to overestimate actual observed counts.71 In that small study of 12 youth females over a 20-game season, the authors concluded ’all players overestimated the number of headers compared with game video. Linear regression analysis indicated that self-reported headers overestimated the actual number of headers counted by 51%'.

Neuropsychological

Conclusive statements regarding whether there are either acute or chronic neurocognitive effects from heading are elusive because studies to date have used different research methodologies, populations, neuropsychological tests, testing paradigms, outcome measures, clinical and psychometric change criteria. Additionally, it is difficult to isolate the individual effects of heading, concussion, the interactions among concussion and heading and concussion and athletic exposure.

Studies have shown no effect of heading on cognitive functions after a 15 min heading session,80 after two practices or games,81 82 following one season,78 79 83 84 two seasons85 or when assessed as part of a cross-sectional analysis.86 Conversely, some older studies have shown minor cumulative effects of heading on cognition.87 88 A recent systematic review of 22 studies, including 2228 participants, 42% female, age 13–70 years found ‘no overall effect for heading a football on adverse outcomes'.89

Although limited, the preponderance of available data in adolescents and adults does not support the hypothesis that there are intermediate or long-term adverse neurocognitive effects from heading the ball in soccer. Any adverse short-term effects appear to be limited to short-lived transient (<48 hours postinjury) changes in postural control,90 near point of convergence91 92 and headache.93

Biomarkers

Maher et al 94 reviewed the data related to the biomarkers and neurocognitive outcomes of acute and chronic heading exposure. In terms of short-term effects, Maher et al found increased but reversible changes in brain biomarkers S100-B and NSE (neuron-specific enolase) and no consistent evidence of neurocognitive deficits, the latter likely due to lack of experimental control for player history (eg, number of prior concussions, playing in the leather ball era, learning disabilities, alcohol intake) and the interaction of heading exposure and concussion history. Regarding the long-term effects of heading, Maher et al reported the following: a) no evidence of residual biomarker abnormalities, b) ‘the possibility’ of lasting deficits in neurocognitive functioning in those with a history of concussion and c) significant differences in neuroimaging or neurophysiological results in five of six studies that compared soccer players and non-playing healthy controls. Regarding the neuroimaging findings, however, Maher et al concluded that the studies were ’limited—particularly in terms of acute concussion where they are lacking completely—and, as such, it is an area of research that should be expanded on in the future to determine whether there truly are adverse effects related to concussions and heading'.94

Future research regarding potential effects of heading and soccer participation needs to address host and biopsychosocial variables such as age, sex, other sport and recreational head trauma exposure, personal and family medical and psychiatric histories, learning disabilities (including attention deficits), migraine headaches and other health conditions. The measures used to assess cognitive domains should be harmonised along with the timing of neurocognitive testing and the criteria used for assessing meaningful change. Consideration should also be given to using broader (vs screening) measures of cognitive performance (eg, full neuropsychological evaluations, academic performance, standardised test scores).

Neurodegenerative disease and chronic traumatic encephalopathy

With the increased attention to CTE, there has been developing interest in examining whether a relationship exists between sports participation and the development of CTE pathology or neurodegenerative diseases and, if a relationship does exist, identifying the nature and strength of that relationship.

According to provisional consensus guidelines published in 2016, CTE pathology is characterised by abnormal accumulation of hyperphosphorylated tau protein (P-tau) within neurons, astrocytes and cell processes around small blood vessels in an irregular pattern at the depths of cortical sulci. Interpretation of immunohistochemical stains for P-tau using autopsy tissue is required to make this assessment. P-tau is also a secondary accumulation in many conditions, including Alzheimer’s disease, Down syndrome, prion diseases, neurodegeneration with brain iron accumulation type 1, postencephalitic parkinsonism and Niemann-Pick disease type C, among other conditions. P-tau is almost universally detected at autopsy with advanced age. Although there is work in progress,95 no accepted criteria for diagnosing CTE in a living person exist at present. Consequently, the incidence of CTE pathology in the general population is unknown.

The data regarding CTE pathology in soccer are limited to a small number of case studies, or retrospective analyses of brain bank specimens.96–103 A recent paper by Zuckerman et al has reviewed the data for CTE in contact (soccer, rugby, ice hockey) and collision (American football) sports.104 Sulcal depth p-tau consistent with the solitary necessary and sufficient CTE criterion was described in about half of the subjects studied, all of whom had comorbid neurodegenerative diseases (eg, Alzheimer’s disease, frontotemporal dementias, amyotrophic lateral sclerosis, etc), which makes it difficult to attribute disease onset and progression solely to CTE pathology. Current research in this area is focusing on the effects of cumulative subconcussive impacts.105

It is expected that former soccer players will develop neurodegenerative diseases at the same rates as found in the general population. At present, no data exist that support the hypothesis that soccer participation is a risk factor for the development of neurodegenerative disease. Similarly, at this time, consistent with evidence discussed in the CISG Berlin,1 106 there are no data to support a causal relationship between soccer and CTE pathology. Continuing research in this area is a necessity to more fully understand whether a link does exist, and if such a link does exist, whether there are approaches for mitigating such risk.

The use of headgear

Hard shelled sport helmets are designed to prevent catastrophic injury (eg, skull fracture, intracranial hematomas).108 They are not currently designed to prevent concussion. In contrast, soft shelled headgear has been developed under the hypothesis of preventing concussion through the use of energy absorbent padding designed to reduce head impact force and subsequent acceleration. Laboratory testing has shown that headgear is effective in reducing impact force at high speeds in simulated head-to-head impacts.109–114 Headgear was not found to be effective, however, in reducing acceleration forces during heading.114 Testing in the laboratory has also demonstrated that sex differences exist with headgear increasing head impact acceleration in females when compared with males as well as when compared with controls (no headgear).115

Other studies have identified that headgear is ineffective at mitigating subtle neurocognitive effects of an acute bout of soccer heading.116 In the CISG systematic review on prevention, the conclusion was ‘the literature regarding the use of soft shelled protective headgear in sport is mixed'.117 The effect of soccer headgear on reducing concussion incidence is unknown as there is a paucity of high-quality field research to date. Soccer headgear can potentially reduce impact acceleration during stiff impacts, such as head-to-head, elbow-to-head, head-to-goalpost or head-to-ground impacts. There is no support for using headgear to decrease the impacts of heading, and in females the addition of headgear may actually increase the forces that the head experiences.115 Wearing headgear may also change athlete playing behaviour, giving them a false sense of protection and potentially playing more aggressively (eg, gladiator effect or ‘risk compensation’).118 119 This is an area where further research is necessary.

Restrictions on heading

Prevention measures have been instituted in other sports to limit exposure or risk. For example, baseball has moved towards avoiding collisions at home plate, ice hockey has focused on avoiding contact and checking in youth play and football has implemented various measures to limit or avoid contact exposures in youth play. For soccer, although it is clear that heading in and of itself is not associated with significant acute or chronic dysfunction, a common mechanism of SRC being a blow to the head (from another head, elbow, knee or ground) during or after an aerial challenge. This is the basis for the restrictions of heading that were imposed for American youth soccer by US Soccer.

When exploring the issue of limiting heading in specific age groups, it is important to differentiate between restriction (which means to limit) and a ban (which means to prohibit). US Soccer’s current policy49 on age-related heading prohibits heading for 10 and under (U-11). Youth between the ages of 11 and 12 years (U-12, U-13) may engage in heading training for ‘a maximum of 30 min per week with no more than 15–20 headers per player, per week’. There are no restrictions for U-14 and older. At present, there are no studies that have empirically examined the effects of age-dependent restrictions on heading.

The US Soccer DA has also implemented additional safety components, which affect approximately 17 000 players who participate in the DA. Examples of safety measures that have been incorporated into DA play include small-sided games of 7–9 vs 7–9 (on a smaller field), where there is an emphasis on ball control that minimises long punts and aerial challenges. In addition, these small-sided games promote playing the ball out of the back on the ground (vs a goalkeeper punting the ball), and additional rules that do not allow the goalkeeper to punt the ball or use a drop kick, which may be a consideration in the future.120

Other modifications to game play can include modifying the field size, the number of players, whether to use rules such as off-sides, and other rule changes. Additionally, Athletic Trainers (ATC) in the DA are mandated to be available and present for every DA match. Finally, in suspected head injuries, injury substitutions are allowed (that do not count against the team’s number of allowed substitutions) and sufficient time is provided for the player to be evaluated by the ATC in real time and determine return to play status. If the player is medically cleared for return to that match, the player replaces the original substitute.49

Strength training for the neck

It seems logical that strengthening the neck muscles should be protective against head injury. The current data are, however, inconsistent. Mihalik et al 121 showed no difference in resultant head accelerations across neck strength tertiles in youth hockey players. In American football, increases in neck stiffness, not strength, reduced the odds of moderate or severe impacts.122 During heading, Dezman et al 123 noted that strength imbalances at the neck contributed to greater head acceleration, Bretzin et al 124 said that neck strength and girth were negatively correlated to linear and angular acceleration and Caccese et al 125 found that neck size and strength explained nearly 25% of the variation in linear and angular acceleration of heading.

Given the observations specific to soccer, improving neck strength should be effective at reducing accelerations during impact. It is well known that the neck musculature is amenable to training126–128 and short-term training (5 weeks, twice per week in professional rugby players) can increase neck strength by nearly 20%,128 but simply increasing neck strength and girth of soccer players has no effect on head stabilisation during heading.129

The effect of improved neck strength on head stability and reduced injury risk is unknown. The High School Reporting Information Online project collected anthropometric, strength and injury records from over 6000 high school athletes. The data showed that smaller girth (as neck-to-head ratio) and lower neck strength (hand-held dynamometry) were related to increased incidence of concussion. The authors suggested that every one pound increase neck strength was accompanied by a 5% decrease in the odds of a concussion.130 Despite some promising, and conflicting, results, a randomised clinical trial of neck strengthening and subsequent concussive injury remains to be conducted.

Laws of the Game

The Laws of the Game (Law 12. Fouls and Misconduct) do not single out any specific injury. In the leadup to the 2006 FIFA World Cup, the IFAB was presented with an analysis of head injury mechanisms that prompted them to grant the referee the authority to punish intentional elbows to the head with an immediate red card.

After giving the referee the authority to sanction intentional elbows to the head, a study of Norwegian male professional football players,29 reported that strict enforcement and interpretation of the Laws of the Game was indeed associated with a lower incidence of head incidents caused by player-to-player contact and lower incidence of arm-to-head contact. In a 13-year period (before and after the IFAB ruling), all head and facial injuries were reduced in the German professional league. Specifically, concussions have been reduced by 29%.131 Both studies indicate that enforcement of existing rules can have an effect on foul play and limit concussive injury.

For the 2014 FIFA World Cup, FIFA’s Sports Medical Committee proposed that in cases of suspected concussion, the referee will have the authority to stop the game for 3 min to allow for an on-field of evaluation of the injured player. The referee will only allow the player to continue with the authorisation of the team doctor. In 2016, US Soccer modified its substitution rule for the Academy programme as follows: ’any player suspected of having head injury may be substituted for evaluation without that substitution counting against the team’s total number of allowed substitutions during the game'.49

The properties of the ball as well as the playing surface are areas that potentially have implications for injury prevention. There are guidelines for ball size (specified in Law 2) that can affect the forces that the head experiences. It is important that players, parents and coaches use the appropriate sized ball for youth players and attend to the inflation pressure of the ball.132 Whether different playing surfaces are associated with different risks for injury is at this time unclear, but is an area of future research.117 Finally, an area of significant interest is in awareness training that might increase an athlete’s ability to prepare and brace for an impact to the body or head, which might include aerial challenge training for the soccer player.68 133 This is an important area of future research.

Educational efforts

Educational efforts by governing bodies (eg, US Soccer, US Youth Soccer, NCAA, National Federation of High Schools) and government (Centers for Disease Control and Prevention) are important and target athletes, parents, coaches and healthcare providers in terms of the importance of early recognition and management of SRC.134–136 US Soccer has produced an educational video on concussions and has implemented this video for coaching and referee certification as well as player and parent education and awareness (www.RecognizeToRecover/Head-and-brain). Finally, the DA has implemented a pledge to concussion awareness by all stakeholders to build a culture of safety around this injury. MLS has also created an internal concussion education video that is shown to all players and coaches in the preseason. Internal MLS data suggest that the video is effective in increasing knowledge on concussion. MLS has also created concussion education posters that are placed in all in team locker rooms.