As many as 3.8 million sport-related traumatic brain injuries may occur annually in the United States (13), and the incidence between sports widely varies. Fortunately, most of these injuries are classified as mild traumatic brain injuries (mTBI) or concussions, yet the incidence in some sports (e.g., women's ice hockey) may exceed that of all other injuries sustained in a season (1). The U.S. Centers for Disease Control and Prevention also report a high incidence of recurrent concussions in several sports such as football and warns that the likelihood of serious sequelae increases with repeated injury (2). Our earlier research has indicated that a history of previous concussion increases the risk for future concussion, as well as the acute outcomes after subsequent concussion (10,13). Other factors such as learning disabilities (3) and age (6) also have been shown to influence the risk for and outcomes after concussion. Research on the influence of various biomechanical factors for predicting outcomes after sport-related concussion is inconclusive, but new technologies may lead us to more answers.
The biomechanics of head impacts have been investigated in a variety of laboratory settings over the past six decades. Ommaya and Gennarelli (25) were among the first to describe in detail linear and rotational accelerative mechanisms of injury using animal models that helped to better explain the role of linear versus rotational acceleration for brain injury. More recently, the National Football League (NFL) commissioned an investigation of sport-related concussions. One of these studies performed laboratory reconstructions of video-recorded concussions using helmeted Hybrid III dummies (28), and it was suggested that an injury threshold of 70g-75g existed for sustaining concussion based on the translational (linear) acceleration of a football player's head.
Linear and rotational head accelerations are hypothesized to be the primary risk factors for concussion during an impact. Both direct and inertial (i.e., whiplash) loading of the head may result in linear and rotational head acceleration. Head acceleration induces strain patterns in brain tissue, which may cause injury. Current science has not identified an exact threshold for concussive injury, and direct measurement of brain dynamics during impact is extremely difficult in humans. Head acceleration, on the other hand, can be more readily measured; its relationship to severe brain injury has been postulated and tested for more than 50 yr. Both linear and rotational accelerations of the head play important roles in producing diffuse injuries to the brain. However, the relative contributions of these accelerations to specific injury mechanisms have not been conclusively established. The numerous mechanisms theorized to result in brain injury have been evaluated in cadaveric and animal, surrogate (26), and computer (34,35) models. Prospective clinical studies combining head impact biomechanics and clinical outcomes have been strongly urged but have been relatively void in the literature.
Concussions are often referred to structurally as "diffuse axonal injuries" and result in some degree of functional impairment but differ from more moderate to severe TBI in that the impairment is transient in nature. Diffuse axonal injury, in addition to linear coup-contrecoup mechanisms of injury, can result in disruption to centers of the brain responsible for breathing, heart rate, and consciousness, but more typically result in memory loss, cognitive deficits, balance disturbances, and a host of other somatic symptoms. Our prior work has identified recovery curves for symptoms, cognitive function, and balance, with deficits typically lasting 7-10 d after concussion in high school and college athletes (12,18). Reflection upon these findings has led to the question, What is the relationship between clinical outcome measures from our earlier work and biomechanical factors? The literature has not adequately addressed this question. We hypothesize that within the spectrum of concussion or mTBI, the biomechanical threshold for sustaining the injury is not only elusive, but impact severity (measured in acceleration/deceleration) may be clinically irrelevant. This review aims to collate findings from our recently conducted studies investigating biomechanical relationships with various factors such as playing position, types of play, concussive versus subconcussive impacts, location of impacts, and clinical measures of concussion.
THE BIOMECHANICS OF CONCUSSION AND IMPACT TO THE BRAIN
The biomechanics of TBI remains an area elusive to many researchers. Investigators in this area face a number of challenges in trying to understand head injury impact mechanics. Current ethics standards have made the use of primate and other mammalian animal models very difficult to pursue; animal basic research in this area has been limited to the rat and small mammals in recent years. Second, the use of postmortem cadavers does not allow researchers the ability to study impact mechanics in the context of everyday activities, including sport participation and work. The lack of muscle tonus and decreased volumes of cerebrospinal fluid further make it difficult to replicate an in vivo sample in the context of this area of study.
In the context of mTBI, the term "impact" typically denotes an injurious blow that makes direct contact with the head. An indirect impact, on the other hand, refers to an impact that sets the head in motion without directly striking it. Examples of impacts range from helmet-to-helmet collisions, striking an opponent's head with a stick, or being struck in the head by a projectile used in the sport (e.g., soccer ball, hockey puck, etc.). Indirect impacts are most commonly caused by tackling or body checking and are the result of abruptly stopping an opponent's body from traveling in the direction in which it was headed. To relate this notion in laypersons' terms, it is similar to the effect experienced by passengers when a car quickly accelerates or stops. Direct and indirect impacts are traditionally linear (translational) or angular (rotational) in nature. In real-world activities, there are usually some combinations of both linear and angular accelerations associated with direct and indirect impacts. Many factors are thought to play a role in the body's ability to dissipate head impact forces, including individual differences in cerebrospinal fluid levels and function, vulnerability to brain tissue injury, relative musculoskeletal strengths and weaknesses, and the anticipation of an oncoming direct or indirect impact. Thus, the relative contributions of angular and linear accelerations are not clearly understood with respect to mTBI.
Another underlying question in this area of research is why does every direct or indirect impact not result in an injurious episode? If the head does not move after a collision, the kinetic energy transferred by the blow should theoretically be transmitted elsewhere, leaving the athlete otherwise unharmed. It was this principle that did not allow researchers before Denny-Brown and Russell to more fully understand the phenomenon of mTBI (4). However, even when the head does not move, kinetic energy can still be transferred through the skull, resulting in internal, potentially injurious, deformations. Furthermore, cerebrospinal fluid protects the brain within the cranium. As a result, some direct and indirect impacts do not exceed a threshold needed to drive the brain to impact the inside walls of the cranium and cause transient lesions and subsequent mTBI. When an athlete experiences a rotational mechanism, it is thought that rotation of the cerebrum about the brainstem produces shearing and tensile strains. Because activities in the midbrain and upper brainstem are responsible for alertness and responsiveness, rotational mechanisms of TBI are believed to more likely result in loss of consciousness than predominantly linear types of impacts (25). Regardless of the type, attribute, or severity of a particular impact, the end result is as follows: the mass of the head has become too large for the body to overcome the acceleration or deceleration forces that have sent it in motion.
IN-HELMET ACCELEROMETER RESEARCH
Real-time accelerometer data collection is a novel method available to researchers who are attempting to better understand the biomechanics of mTBI, but the earlier study designs were limited and unable to provide a realistic and meaningful interpretation of the data. For example, in a multisport study, Naunheim et al. (23) attempted to investigate the linear accelerations sustained by high school student-athletes - specifically, an ice hockey defenseman, football offensive lineman, football defensive lineman, and a soccer player. A triaxial accelerometer was inserted within a football and ice hockey helmet, and linear acceleration values were recorded during actual play. The data obtained from the soccer player allowed for limited interpretation firstly because there was no method of affixing the accelerometer to the player's head so the soccer player wore an instrumented football helmet. Secondly, game data were not captured; instead, the soccer player was asked to head 23 balls kicked to him or her at a standardized velocity. The mean linear acceleration measured in the football and ice hockey players was 29.2g and 35.0g, respectively.
Duma et al. (5) were the first to use acceleration-measuring technology in helmets for large numbers of athletes during normal practice and game situations. His group used the Head Impact Telemetry (HIT) System technology (Simbex, Lebanon, NH) incorporated within the Sideline Response System (Riddell Corp.; Elyria, OH). A major component of the HIT System is a unit composed of six spring-loaded single-axis accelerometers that are inserted into football helmets (Fig. 1). Duma et al. reported the magnitude of head impacts to be 32 ± 25g. This contrasts the range of 20g to 23g in a similar sample of Division I collegiate football players studied by our University of North Carolina research group using the HIT System technology (19). The football-related data reported by Naunheim et al. (23) also are much higher than our observed football impact data, which averaged 22g of linear acceleration (19). A statistically significant difference was then observed by comparing the linear acceleration of head impacts of football players across three different event types. Head impacts sustained in helmets-only and full-contact practices were significantly higher than those sustained in games or scrimmages. This finding was somewhat surprising, given that our earlier investigations found the incidence of concussion to be 6-8 times higher in games than practices (10,13). Furthermore, our preliminary data in youth ice hockey players suggest that mean linear accelerations average about 19g (22). Although the earlier study by Naunheim et al. (23) represented an important advance toward real-time data collection, her group was limited by a very small sample and did not transform the data to render it to be normally distributed; this tends to overestimate the actual linear acceleration values measured and leads us to believe that the actual values were probably closer to those captured in the more recent studies involving the HIT System.
A number of explanations exist to account for these differences. Linear acceleration is a highly skewed measure, with most of all head impacts yielding low linear acceleration outcomes. Duma et al. (5) calculated the mean and SD of their impacts without first controlling for the highly skewed distribution of their data. Furthermore, they alternated eight accelerometer units among their sample, selectively targeting players throughout the course of the season. The methodology used by our research team involved measuring all head impacts sustained by each study subject (∼50 per season) in every practice and game throughout the season and alternated between players only to replace an individual who no longer remained in our study because of a season-ending injury. Using this methodology, we established that college football players experience approximately 950 subconcussive (without injury) head impacts in a given collegiate season, suggesting that the brain can withstand many impacts in a given season. This is in agreement with work by Schnebel et al. (29), whereby 54,154 head impacts were sustained by 56 football players (40 college and 16 high school; mean of 967 per athlete), with only two reported concussions.
Others have joined in the effort of implementing the HIT System in the realm of high school and college football. It was not until recently, however, that more extensive study of youth ice hockey started. We have been studying head impact biomechanics in Bantam- and Midget-aged youth ice hockey players for two complete hockey seasons. Preliminary data in this sample suggest that 13- and 14-yr-old ice hockey players sustain head impacts nearing the magnitude of those sustained by college football players (22). Although this technology is still considered novel, it is gaining popularity within the research community to better understand the nature of head impacts sustained by athletes. Future research in this regard should focus on helping players to better protect themselves and their opponents from sustaining high-magnitude impacts, which should result in lower incidences of mTBI in amateur, college, and professional sports.
CONCUSSION INJURY THRESHOLD
Whereas our research has provided important insights about recovery after a sport-related concussion (10,12,13,18), we have become increasingly interested in the relationship between the clinical recovery curves and impact biomechanics. Why are some athletes able to withstand very high magnitude impacts without much deficit, whereas others struggle with significantly lower-end impacts? A number of contemporary studies have investigated impact biomechanics and have sought to shed light on proposed injury thresholds for mTBI. In an earlier work, Hodgson et al. (14) imparted short-duration impacts (1-2 ms) to six monkeys and reported that the linear accelerations of the impacts causing concussion ranged from 2000g to 5000g. Unterharnscheidt and Higgens (31) report from their study that rotational accelerations in excess of 200 rad·s−2 produced cerebrovascular injury to most subjects in their sample.
More recently, Gurdjian (8) proposed an mTBI injury threshold in terms of linear acceleration. In their report, they proposed that an mTBI would likely result from a blow to the head exceeding 80g to 90g that was sustained for greater than 4 ms. During the past 7 yr, the NFL has recently published a sequence of studies in Neurosurgery, describing many facets of concussion and mTBI in their league. The initial study from these efforts pertained to the laboratory reconstruction of concussive injuries captured on video (28); they represented the most sophisticated method of analyzing concussive injuries in professional football players at the time. The studies are not without limitations. For one, the laboratory retrospective reenactments were based on game video footage, and important mathematical derivations were extrapolated from relatively low-speed video capture frequencies. Second, only 31 of 182 cases reviewed were reconstructed in the laboratory. Conclusions were made based on this very small and selective sample of cases. Based on these data, Pellman et al. (28) suggest that mTBI in helmeted impacts are likely to occur between 70g and 75g. This contrasts with data we have previously published, where only 7 (<0.38%) of 1858 head impacts exceeding 80g resulted in a diagnosed case of mTBI (19). One reason for the discrepancy may be explained by the fact that Pellman et al. (28) studied professional football players, whereas we investigated this phenomenon in college football players. Given similarities in player size, these differences are unlikely to be explained by the different samples.
Zhang et al. (38) shortly thereafter proposed injury threshold values using the Wayne State University brain injury model. This model replicates a 50th percentile adult male head and includes anatomical structures including the dura, falx cerebri, tentorium and falx cerebelli, cerebrospinal fluid, cerebrum, cerebellum, and brainstem. Before analyses, the model was prevalidated against cadaveric intracranial and ventricular pressure data previously published. Twenty-four head-to-head impacts sustained in professional football were duplicated using their finite element head model to predict injury thresholds based on brain tissue responses. They reported that resultant linear accelerations of the head center of gravity of 66g, 82g, and 106g were associated with a 25%, 50%, and 80% probability of mTBI, respectively. These values are similar to those proposed by Ono et al. (26), who suggested that impacts of 90g sustained for 9 ms or longer would result in mTBI. The Wayne State University Concussion Tolerance Curve, published in 1964 by Gurdjian et al. (9), is a function of impact duration and impact magnitude. It deemed that an 80g impact noninjurious and an impact greater than 90g could produce an mTBI. Zhang et al. (38) also propose rotational accelerations more in line with those we have collected in our own ongoing work. They associate rotational accelerations of 4600 rad·s−2, 5900 rad·s−2, and 7900 rad·s−2, with a 25%, 50%, and 80% probability of sustaining an mTBI. Our University of North Carolina data suggest that there is far from a 50% probability of sustaining an mTBI with an impact exceeding 82g or 5900 rad·s−2 (11,19).
Our work continued in this area, attempting to better understand the effect of sustaining head impacts in excess of previously published injury thresholds. We studied how football players performed on concussion clinical measures after a game or practice session in which they sustained an impact exceeding 90g (17). Athletes were tested only in the absence of a concussion diagnosis within 16-24 h after the session. The most important finding was that nonconcussed football players did not exhibit a decline in balance and cognition after an exposure in which they sustained at least one high impact greater than 90g, which is a proposed theoretical injury threshold. Our findings suggest that clinicians should not expect a single impact greater than 90g to necessarily result in immediate symptoms of a concussion or subsequent balance or cognitive deficits that would suggest the impact affected their overall function 24 h later. Our findings would seem to contradict the notion that a rigid threshold for concussion can be set, given that all 22 players in our high-impact condition sustained impacts well above the proposed threshold of 70g-75g.
THE RELATIONSHIP BETWEEN BIOMECHANICS AND CLINICAL SYMPTOMS
Sport-related concussion typically results from forces directly imparted to the head or indirectly through the neck, resulting in a combination of rapid acceleration and deceleration. Such forces create linear and/or rotational acceleration/deceleration on the brain. The animal literature on this topic suggests that rotational acceleration is more significant than linear acceleration and can lead to more serious effects on the brain (25), but we do not know that this is necessarily the case with sport-related concussion. Concussive injury presents with varying types of symptoms and different levels of symptom severity. This presentation of symptoms also may vary widely depending on the biomechanical forces involved. It has long been understood that the severity of the pathological injury depends on the magnitude, location, and distribution of the forces across the brain tissue, and several research groups have attempted to examine and quantify the accumulated accelerations during sport impacts and how those accelerations are transmitted to the brain of an athlete (24,27,28,30,33,36,37). However, very few researchers have conducted data collection in real time on the field in the environment in which athletes typically encounter concussive impacts and attempted to study the relationship of clinical outcome to the biomechanical measures.
Biomechanical analysis can provide valuable clinical insights into the causes and factors contributing to head loadings and stresses to the brain. Such techniques have involved both empirical and analytical approaches. Empirical methods measure kinematics and kinetics exerted on the body as opposed to analytical methods, which predict bodily responses by replicating the impact dynamics (30). Kinematic analysis techniques provide information on body motions and tend to consist of cinematography and motion-tracking systems providing two- or three-dimensional information on body segments. The kinematic measurement techniques include linear accelerometry for direct measure of head impact response (24). Kinematic measurement generally involves sensors mounted onto the head for direct measurement in one axis or combined with multiple units to quantify accelerations in two or three dimensions. Angular accelerometer techniques also are included in kinematic analysis methods. The angular accelerations about one or more axes are derived from multiple linear acceleration measurements and require stringent implementation to ensure accuracy (37).
Observation and assessment of the biomechanics of injuries in American football also have provided some insight into concussive injury (27,28,32,36). Pellman and colleagues (28) reconstructed head impacts observed in concussion footage. Measures of head acceleration for concussed athletes versus uninjured struck athletes and uninjured striking athletes show a significantly greater rapid change in head velocity for the concussed athlete. Translational acceleration from impacts on the face mask or the side of the helmet or falls on to the back of the helmet was most frequently associated with concussions (28). Helmet-to-helmet contact occurs regularly in American football. Viano and colleagues (32) sought to assess the collision mechanics resulting in injury to the struck player and the biomechanics of the striking players in laboratory tests. A number of translational and rotational head accelerations were examined to determine how the striking player executed the concussive blow. Results revealed the key to the execution of a concussive blow is the head-down position. This occurs when the striking athlete lowers his or her head, bringing his or her head, neck, and torso into alignment, allowing an exertion of maximum force on the struck athlete, whose head and neck resist the impact. The concussion was observed to occur as a result of the greater inertia of the striking athlete behind the impact. The head-down position was found to increase the mass of the striking athlete up to 67% by coupling the torso into the impact and transferring more momentum to the struck athlete (32). Viano and colleagues (32) sought to compare the head impact from a boxing blow with impacts observed in American football. The findings show that the punches inflicted by the boxers had high-impact velocity but lower head injury criteria - a measure of the likelihood of head injury arising from an impact - and translational acceleration than in American football impacts but cause proportionally more rotational acceleration because of a lower effective punch mass (32).
Our ongoing study at the University of North Carolina on injury biomechanics in American football players uses a real-time helmet accelerometer data collection methodology in Division I college football players. Our findings suggest a higher propensity of top-of-the-head impacts and a higher relative risk of concussion for those impacts. In this regard, 6 of 13 concussions occurred from impacts to the top of the head, this is in contrast to four, two, and one concussion occurring to the front, right, and back, respectively (Fig. 2) (11). Our findings suggested that football players are concussed by impacts to the head that occur at a wide range of magnitudes (60.51g-168.71g linear acceleration), and that clinical measures of acute symptom severity, balance, and neuropsychological function all appear to be largely independent of impact magnitude and location. There was no relationship between impact magnitude or location, and clinical outcomes of symptoms, balance, or neuropsychological performance (Table). In short, the concussions sustained as a result of lower end magnitudes tended to present with just as many clinical deficits as those with higher end magnitudes. Thus, despite the literature suggesting that high magnitudes of head impact, particularly with high-angular acceleration, result in more serious clinical outcomes in cases of moderate or severe TBI (15,25), the magnitude and location likely do not predict clinical recovery in cases of mTBI.
The uniqueness of this study was that it combined impact biomechanics captured in real time with clinical measures of symptom severity, neurocognitive function, and balance captured during the acute period after concussive injury. The findings supported the notion that the threshold for mTBI (concussion) is elusive and added further to the debate as to whether it is lower or higher than previously thought. Given the findings of our companion articles (17,19), one could make the argument that, on average, the threshold must be higher than 80g or 90g. Impacts greater than 90g, in the absence of self-reported concussion symptoms, did not result in a diagnosed concussive episode (17) and fewer than a half percent (<0.35%) of all impacts greater than 80g resulted in a diagnosed concussion (19). If the threshold were in the 80g to 90g range for mTBI, one would expect more prevalence of symptoms in impacts of these high-end impacts. Our data also suggest that top-of-helmet impacts may result in larger postural stability deficits after mTBI (11). We speculated that top-of-helmet impacts might result in a coup-contrecoup mechanism occurring in a superior-to-inferior direction, causing the cerebellum to impact the base of the skull and recoil superiorly into the cerebellar tentorium. Interestingly, our data also indicate that top-of-helmet impacts typically result in relatively lower rotational acceleration values compared with injuries after impacts to the other areas of the head. However, we observed that these impacts to the top are at least six times more likely to result in impact magnitudes greater than 80g of linear acceleration than side or front impacts (19). These findings bring into question the notion that rotational acceleration is the leading precursor to injury and are suggestive that the type of acceleration, in combination with impact location, may be a better determinant for both onset and severity of injury.
CAN WE STUDY BIOMECHANICS AS A MEANS OF INJURY PREVENTION?
Our youth hockey study suggests that hockey infractions (penalties) occur at a relatively high frequency and typically result in higher measures of head impact severity than legal collisions. Infractions were observed in 17.3% (115 of 665) of all body collisions (21). Collisions involving infractions had slightly higher linear accelerations and HIT severity profiles than collisions with no infraction. The HIT severity profile is a principal component score containing linear acceleration, rotational acceleration, impact duration, and weighted by impact location; it was found to be more predictive of injury than other classical measures of head impact severity alone (7). Elbowing, head contact, and high sticking infractions resulted in greater linear acceleration than collisions with no infraction. A strong trend for higher rotational accelerations in this infraction type compared with legal collisions also was present. We concluded that athletes and coaches should conform to playing rules, and officials should more stringently enforce existing rules and assess more severe penalties to participants who purposefully attempt to foul an opponent at the youth ice hockey level.
In addition, our hockey research findings support the notion that the ability to anticipate a collision may play a role in minimizing head impact severity (20). We found impacts occurring in the open ice and those which were deemed unanticipated resulted in slightly higher impact forces than impacts along the playing boards and those deemed to be anticipated, respectively. This represents a continued need to educate our players with the necessary technical skills needed to heighten their awareness on the ice. Clinically, coaches and athletes should incorporate body-checking exercises in practices and spend time educating young athletes on proper checking techniques to minimize the risk of injury and increase the safety of ice hockey.
These findings, combined with our football results, suggest that the study of biomechanics may be useful in influencing rule changes for improving safety in these respective sports. Such changes would aim to prevent open-field/open ice collisions in which players may be ill prepared and vulnerable to sustain high-level impacts to the head. Future biomechanical studies in both football and hockey are needed to better interpret the findings of an increased likelihood for severe head impacts sustained during top of the head impacts in football and specific plays or conditions in hockey.
Because of the varying magnitudes and locations of impacts resulting in concussion, as well as other factors such as the frequency of subconcussive impacts and number of prior concussions, it may be difficult to establish a threshold for concussive injury that can be applied to football and other helmeted contact sports, such as hockey and lacrosse. As reported previously in the literature, any proposed theoretical injury threshold should be interpreted with caution. Despite this, biomechanics research has still provided us with valuable information for improving safety in sports such as football and hockey. Our findings further substantiate the notion that concussions must be managed using a multifaceted approach (Fig. 3).
The most important findings of these combined studies have been 1) that concussions can occur at lower impact magnitudes than previously thought; 2) that measures of linear acceleration appear equally important to cause concussion as angular acceleration; 3) that athletes can sustain a high number of head impacts in a season (many exceeding 80g-90g) and never sustain a diagnosed concussion; and 4) clinicians should not attempt to use impact magnitude or location to predict acute clinical outcomes of symptom severity, neuropsychological function, and balance. Our earlier studies, combined with those of several other studies on this topic, call for more research to be conducted to investigate how linear and rotational accelerations relate to measures of symptom severity, neurocognitive function, and postural stability in larger sample sizes across the entire recovery period. In addition, the role of this technology should be further investigated to identify its use for behavior modification and improved player mechanics.
This work was supported by the U.S. Centers for Disease Control and Prevention, the Ontario Neurotrauma Foundation, the U.S.A. Hockey Foundation, and the National Operating Committee on Standards for Athletic Equipment. Because of referencing limitations set by Exercise and Sport Sciences Reviews, the authors were limited in their ability to reference many other important works.
1. Agel J, Dick R, Nelson B, Marshall SW, Dompier TP. Descriptive epidemiology of collegiate women's ice hockey injuries: National Collegiate Athletic Association Injury Surveillance System, 2000-2001 through 2003-2004. J. Athl. Train.
2. Centers for Disease Control and Prevention. Sports-related recurrent brain injuries - United States. M.M.W.R. Morb. Mortal. Wkly. Rep.
3. Collins MW, Grindel SH, Lovell MR, et al
. Relationship between concussion and neuropsychological performance in college football players. JAMA.
4. Denny-Brown D, Russell WR. Experimental cerebral concussion. Brain.
5. Duma SM, Manoogian SJ, Bussone WR, et al
. Analysis of real-time head accelerations in collegiate football players. Clin. J. Sport Med.
6. Field M, Collins MW, Lovell MR, Maroon J. Does age play a role in recovery from sports-related concussion? A comparison of high school and collegiate athletes. J. Pediatr.
7. Greenwald RM, Gwin JT, Chu JJ, Crisco JJ. Head impact severity measures for evaluating mild traumatic brain injury risk exposure. Neurosurgery.
8. Gurdjian ES. Prevention and mitigation of injuries. Clinical neurosurgery
9. Gurdjian ES, Lissner HR, Hodgson VR, Patrick LM. Mechanism of head injury. Clinical neurosurgery
10. Guskiewicz KM, McCrea M, Marshall SW, et al
. Cumulative effects associated with recurrent concussion in collegiate football players: the NCAA Concussion Study. JAMA.
11. Guskiewicz KM, Mihalik JP, Shankar V, et al
. Measurement of head impacts in collegiate football players: relationship between head impact biomechanics and acute clinical outcome after concussion. Neurosurgery.
2007; 61(6):1244-52; [discussion 52-3].
12. Guskiewicz KM, Ross SE, Marshall SW. Postural stability and neuropsychological deficits after concussion in collegiate athletes. J. Athl. Train.
13. Guskiewicz KM, Weaver NL, Padua DA, Garrett WE. Epidemiology of concussion in collegiate and high school football players. Am. J. Sports Med.
14. Hodgson VR, Thomas LM, Khalil TB. The role of impact location in reversible cerebral concussion. Proceedings of the 27th Stapp Car Crash Conference. Warrendale, PA: Society of Automotive Engineers, Inc.; 1983:225-240.
15. Holbourn AHS. The mechanics of brain injuries. Br. Med. Bull.
16. Langlois JA, Rutland-Brown W, Wald MM. The epidemiology and impact of traumatic brain injury: a brief overview. J. Head Trauma Rehabil.
17. McCaffrey MA, Mihalik JP, Crowell DH, Shields EW, Guskiewicz KM. Measurement of head impacts in collegiate football players: clinical measures of concussion after high- and low-magnitude impacts. Neurosurgery.
18. McCrea M, Guskiewicz KM, Marshall SW, et al
. Acute effects and recovery time following concussion in collegiate football players: the NCAA Concussion Study. JAMA.
19. 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.
20. Mihalik JP, Blackburn JT, Greenwald RM, Cantu RC, Marshall SW, Guskiewicz KM. Collision type and player anticipation affect head impact severity among youth ice hockey players. Pediatrics.
21. Mihalik JP, Greenwald RM, Blackburn JT, Cantu RC, Marshall SW, Guskiewicz KM. The effect of infraction type on head impact severity in youth ice hockey. Med. Sci. Sports Exerc.
22. Mihalik JP, Guskiewicz KM, Jeffries JA, Greenwald RM, Marshall SW. Characteristics of head impacts sustained by youth ice hockey players. Proc. Inst. Mech. Eng., P. J. Sports Eng. Technol.
23. Naunheim RS, Standeven J, Richter C, Lewis LM. Comparison of impact data in hockey, football, and soccer. J. Trauma.
24. Newman JA, Beusenberg MC, Shewchenko N, Withnall C, Fournier E. Verification of biomechanical methods employed in a comprehensive study of mild traumatic brain injury and the effectiveness of American football helmets. J. Biomech.
25. Ommaya AK, Gennarelli TA. Cerebral concussion and traumatic unconsciousness. Correlation of experimental and clinical observations of blunt head injuries. Brain.
26. Ono K, Kanno M. Influences of the physical parameters on the risk to neck injuries in low-impact speed rear-end collisions. Accid. Anal. Prev.
27. Pellman EJ, Viano DC, Tucker AM, Casson IR. Concussion in professional football: location and direction of helmet impacts. Neurosurgery.
28. Pellman EJ, Viano DC, Tucker AM, Casson IR, Waeckerle JF. Concussion in professional football: reconstruction of game impacts and injuries. Neurosurgery.
29. 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; [discussion 5-6].
30. Shewchenko N, Withnall C, Keown M, Gittens R, Dvorak J. Heading in football. Part 1: Development of biomechanical methods to investigate head response. Br. J. Sports Med.
2005; 39(Suppl. 1):i10-25.
31. Unterharnscheidt F, Higgens LS. Pathomorphology of experimental head injury due to rotational acceleration. Acta. Neuropathol.
32. Viano DC, Casson IR, Pellman EJ, et al
. Concussion in professional football: comparison with boxing head impacts. Neurosurgery.
33. Viano DC, Casson IR, Pellman EJ, Zhang L, King AI, Yang KH. Concussion in professional football: brain responses by finite element analysis. Neurosurgery.
34. Viano DC, King AI, Melvin JW, Weber K. Injury biomechanics research: an essential element in the prevention of trauma. J Biomech.
35. Viano DC, Lovsund P. Biomechanics of the brain and spinal cord: analysis of neurophysiological experiments. Crash Prev. Inj. Control.
36. Viano DC, Pellman EJ. Concussion in professional football: biomechanics of the striking player. Neurosurgery.
37. Withnall C, Shewchenko N, Gittens R, Dvorak J. Biomechanical investigation of head impacts in football. Br. J. Clin. Psychology.
38. Zhang L, Yang KH, King AI. A proposed injury threshold for mild traumatic brain injury. J. Biomech. Eng.