LITERATURE REVIEW J Neurosurg 126:768–781, 2017

Helmet efficacy against concussion and traumatic brain injury: a review

Je Yeong Sone,1 Douglas Kondziolka, MD, MSc,1 Jason H. Huang, MD,2 and Uzma Samadani, MD, PhD3

1Department of Neurosurgery, New York University School of Medicine, New York, New York; 2Department of Neurosurgery, Baylor Scott & White Central Division, Temple, Texas; and 3Department of Neurosurgery, Hennepin County Medical Center, University of Minnesota, Minneapolis, Minnesota

Helmets are one of the earliest and most enduring methods of personal protection in human civilization. Although primar- ily developed for combat purposes in ancient times, modern helmets have become highly diversified to sports, recre- ation, and transportation. History and the scientific literature exhibit that helmets continue to be the primary and most effective prevention method against traumatic brain injury (TBI), which presents high mortality and morbidity rates in the US. The neurosurgical and neurotrauma literature on helmets and TBI indicate that helmets provide effectual protection against moderate to severe head trauma resulting in severe disability or death. However, there is a dearth of scientific data on helmet efficacy against concussion in both civilian and military aspects. The objective of this literature review was to explore the historical evolution of helmets, consider the effectiveness of helmets in protecting against severe intracranial injuries, and examine recent evidence on helmet efficacy against concussion. It was also the goal of this report to emphasize the need for more research on helmet efficacy with improved experimental design and quantitative standardization of assessments for concussion and TBI, and to promote expanded involvement of neurosurgery in study- ing the quantitative diagnostics of concussion and TBI. Recent evidence summarized by this literature review suggests that helmeted patients do not have better relative clinical outcome and protection against concussion than unhelmeted patients. https://thejns.org/doi/abs/10.3171/2016.2.JNS151972 KEY WORDS helmet efficacy; concussion; traumatic brain injury; review

he most prevalent method of preventing or mini- city Umma after a border dispute and commissioned the mizing traumatic brain injury (TBI) is helmet use. Stele of the Vultures to record his triumph (http://www. Helmets are one of the earliest and most enduring louvre.fr/en/oeuvre-notices/stele-vultures). The upper methodsT of personal protection in the history of human registers of the stele depict helmeted soldiers carrying civilization. They provide protection from head trauma by spears and marching in formation behind their king, Ean- absorbing the impact energy and diffusing and displacing natum.109 These helmets were composed of copper and peak impact and pressure gradient to a greater surface area were most likely padded with leather underneath, as is of the head, rather than to a localized region.76 evident from the bodies excavated from the Royal Cem- etery at Ur, now in modern-day Iraq.113 The Standard of Ur from the Royal Cemetery also depicts similar helmets to History of Pre-20th–Century Helmets those that appear on the stele (http://www.britishmuseum. Since the helmet’s inception, the vast majority have org/research/collection_online/collection_object_details. been designed for ceremonial uses and for the purpose of aspx?objectId=368264&partId=1). Sumerian helmets in- personal protection during military engagements. Early dicated the first major advancement in personal defense in military helmets were used by Sumerians in Mesopota- military conflicts. mia around 2500 bce.113 Circa 2450 bce, Eannatum, king Since then, helmets have been used throughout the of the Sumerian city Lagash, defeated the neighboring Bronze Age, the Roman Empire, the World Wars, and

ABBREVIATIONS ACH = Advanced Combat Helmet; ACRM = American Congress of Rehabilitation Medicine; CTE = chronic traumatic encephalopathy; DAI = diffuse axonal injury; GCS = Glasgow Coma Scale; ICH = intracranial hemorrhage; ICP = intracranial pressure; ImPACT = Immediate Post-Concussion Assessment and Cognitive Testing; LOC = loss of consciousness; LOS = length of stay; mTBI = mild traumatic brain injury; RSC = referee stop contest; SAH = subarachnoid hemorrhage; SCAT3 = Sport Concussion Assessment Tool, 3rd edition; SDH = subdural hematoma; SRC = sports-related concussion; T-tau = total tau. SUBMITTED August 22, 2015. ACCEPTED February 23, 2016. INCLUDE WHEN CITING Published online May 27, 2016; DOI: 10.3171/2016.2.JNS151972.

768 J Neurosurg Volume 126 • March 2017 ©AANS, 2017

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC Literature review of helmet efficacy against head injury presently. Ancient combat helmets were typically made of stiffened leather, bronze, or iron. Notable examples include the bronze helmets of the ancient Greeks (Fig. 1). Steel came into prevalent use in helmets of later eras, especially those of European origin (Fig. 2). There are countless ac- counts of helmet use in battle throughout human history, but literature specifically written on helmet protection against TBI in both civilian and military spheres did not come into prevalence until the 19th century. According to a book published by Jeffreys in 1858, a new helmet design was presented for protection of British soldiers against the sun and TBI.46 The book asserted that the helmet design “would render it proof against injury from ordinary acci- dents, as falls or pressure” and provided “protection from sword-cuts.”46 The Tuson helmet was composed of a steel wire frame padded with cork and cloth.36,46 In 1889, in the British Medical Journal, Foulis corroborated that the Tuson helmet design is “to a great degree, swordproof.”36 Erichsen wrote a case report in 1875 about a lieutenant colonel who fell from horseback and hit his head on the hard road on September 4, 1872, near Poonah, India.35 Ac- cording to the report, the lieutenant colonel’s skull was not fractured due to his helmet, but he was temporarily paralyzed. Another case report in the American Journal of the Medical Sciences, published in 1882, describes a head injury patient who was hit over the parietal bone by a shot from an arquebus firearm while wearing a helmet.54 The report states that the ball indented the helmet and did not penetrate the scalp, but the patient died after 6 days of FIG. 1. Ancient Greek bronze helmet of Chalcidian type from South Italy, coma. circa 350–300 bce. Digital image courtesy of the Getty’s Open Content Program. J. Paul Getty Museum. Public domain. Figure is available in History of the 20th Century and Modern color online only. Helmets Military and civilian helmets in the 1940s were typi- which the focus has shifted from protection against lethal, penetrating TBI toward protection against the more fre- cally composed of a solid steel shell and a plastic and 50,109 cotton fiber liner. Steel M1 helmets were used by the US quently incurred blast injury and concussion. Armed Forces during World War II in both the European and Pacific Theaters (Fig. 3). The surgical and neuro- Protection Against Severe Injuries surgical literature during World War II reported that the Literature evidence relevant to TBI clearly demonstrates government-issued steel helmets reduced the impact of that helmet use provides effectual protection against se- TBI and allowed soldiers to withstand certain blows to the 18,34,45 vere and lethal intracranial injuries like penetrating TBI head that otherwise would have been fatal. Studies that require neurosurgical intervention.22–24,37,84,102 In the on civilian head trauma during the 1940s examine the ef- 17,19 Vietnam War, although combat helmets were not effec- ficacy of motorcycle crash helmets against TBI. One tive against bullets, they provided significant protection study of 106 patients concludes that crash helmets are ef- 20 19 against shell fragments. Improvements in helmet design fective in reducing localized head trauma. Another study and the development of sophisticated fiber materials like in which 8 cases were observed advocates the use of crash Kevlar by the time of the Iraq and Afghanistan Wars re- helmets but states that there is no conclusive evidence in 17 duced the frequencies of penetrating TBI among active- favor of motorcycle crash helmets. duty soldiers.84,93 Retrospective studies on the Iraq and Af- Modern combat helmets generally use a combination ghanistan campaigns provide evidence that Kevlar helmets of synthetic polymers and fibers that provide better protec- were efficacious against penetrating head injuries.93,114 tion than steel shells such as the M1 helmet. This is most Studies on motorcyclists and bicyclists also demonstrate evident in the Advanced Combat Helmet (ACH) of the US that helmet use was associated with reduced frequency of Armed Forces (Fig. 4). The ACH is composed of a Kevlar sustaining severe TBI in a vehicular accident.44 Helmet ef- and phenolic-resin composite shell and a polyurea suspen- 40 ficacy against severe and penetrating TBI in both civilian sion padding system. The ACH provides some degree and military aspects is generally well accepted. of protection against mild TBI (mTBI).40 Modern helmets for civilian purposes, such as cycling, also use synthetic polymers, foam, and fibers for better protection against Protection Against Concussion head trauma (Fig. 5). There has been a general trend in Although the general consensus is to encourage the use the modern era for both civilian and military helmets, in of helmets for head protection, it is unclear according to

J Neurosurg Volume 126 • March 2017 769

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC J. Y. Sone et al. the literature whether helmets are effective against con- cussion.8,103 There is a general lack of published studies specifically on the relationship between helmets and con- cussion. Studies that have provided data on helmet protec- tion from head trauma have focused on head injury as one aggregate group, confounding results specifically about concussion.112 The literature on head trauma that focuses on helmet protection either has found that the data are in- conclusive or has provided disparate conclusions.8,28 Optimization of helmets to maximize protection against concussion is complicated by the limited understanding on the biomechanics of concussion. There is significant evidence in the literature that concussion is primarily caused by inertial forces, or acceleration.75 However, little research and understanding have been developed on the specific roles of linear and angular head acceleration in concussion.72 More recent literature has provided evidence that angular, or rotational, acceleration plays a large role in the biomechanics of concussion.28,69 Furthermore, ex- tensive research has been conducted on the biomechan- ics of impact-induced TBI, but not on blast-induced TBI.78 The protective mechanism of helmets, especially combat helmets, in response to blast events is relatively unknown in the literature.50 Another factor contributing to the com- plexity of the data on helmet protection from concussion is the varying, inconsistent standard for helmet development. It is difficult to balance the needs for maximum protection against concussion with appropriate design for the specific sport or use, affordability, low weight, and comfortable fit, and still develop an effective helmet.50 FIG. 2. Spanish Morion helmet made of engraved and embossed metal, circa late 1600 ce, public domain artifact and image from the Oakland Museum of California. Public domain via CC0 (https://creativecommons. org/publicdomain/zero/1.0/deed.en). Figure is available in color online Epidemiology of TBI only. Maximizing helmet efficacy against concussion and other forms of TBI is important because TBI is a major TBI die soon after the injury, and many patients who sur- public health issue in the US, with very high occurrence, vive and recover from TBI still experience reduced life ex- morbidity, and mortality rates.29,53,106,107 Approximately 1.7 pectancy and a lifetime of neurocognitive deficits.88,106,107 million people in the US sustain TBI annually, in which there are 275,000 hospitalizations and 52,000 fatalities.53,88 Traumatic brain injury is a contributing factor to 30.5% of Clinical Definitions of TBI all injury-related deaths,53 and it is also the leading cause There are various forms and severities of TBI. These of mortality and disability in young individuals in devel- injuries can be categorized into closed TBI (blunt force oped countries.59 There is a recent shift in the incidence trauma), penetrating (open) TBI, and blast TBI.1,59 The se- of TBI toward older individuals, with falls as the primary verity of TBI has historically been most frequently evalu- cause of TBI.79,88 Hospitalizations and mortality rates due ated with the Glasgow Coma Scale (GCS).1,59 Mild TBI is to TBI incurred in geriatric patients by falling have risen defined as a GCS score of 13–15, moderate TBI is defined since 2000 and are expected to continue to increase.41,79 as a GCS score of 9–12, and severe TBI is defined as a The epidemiology of pediatric TBI in North America, Eu- GCS score of 8 or less.1 Another way in which TBI is dis- rope, Australia, and New Zealand indicates that the lead- tinguished is by the location of the intracranial injury. The ing mechanism of injury in children under the age of 5 injury can be extraaxial, such as with epidural hematoma years is also by falling, whereas the mechanism of injury and subdural hematoma (SDH) or hemorrhage, or the in- in children over the age of 15 years is by motor vehicle jury can be intraaxial.1 Primary injury in TBI is neuronal accidents.104 The annual mortality rates of moderate to se- damage at the moment of impact, whereas secondary in- vere TBI, mTBI, and head injury without TBI are approxi- jury is neuronal damage caused by consequent physiologi- mately 6.7%, 1.4%, and 1.9%, respectively.107 Traumatic cal response to the initial injury.1 Acute severe TBI with brain injury also accounts for up to $80 billion annually in intracranial hemorrhage (ICH) or intracerebral contusions costs of care.25,29 The incidence of TBI is rising globally, may lead to death. primarily due to the increase in injuries associated with The most common type of TBI is concussion, which motor vehicles.88 Males have a higher rate of sustaining accounts for up to 75% of all TBI cases.29,38 The term con- TBI than females in every age group—sometimes up to 3 cussion has often been used interchangeably with mTBI times the proportion of females in all TBI cases.53,59,106 A in various publications, but there is a recent trend to rec- significant percentage of patients who suffer from acute ognize concussion and mTBI as different injuries.33,64,65,100

770 J Neurosurg Volume 126 • March 2017

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC Literature review of helmet efficacy against head injury

FIG. 3. Steel M1 helmets worn by US Army soldiers approaching Omaha Beach on D-Day, June 6, 1944. Photography by Conseil Régional de Basse-Normandie, National Archives USA. Public domain.

There is little consensus on the definition and diagnosis tal state at the time of injury as characterized by disorien- standards of concussion or mTBI, and it has also been tation and confusion.88,91 The ACRM definition also states found that highly trained physicians often disagree on the that the focal neurological signs of mTBI may or may not diagnosis of blast-related mTBI.13,110 The biomechanics of be transient, whereas the WHO definition of mTBI de- concussion are also disputed and very little understood. scribes the focal neurological signs as transient.91 The American Congress of Rehabilitation Medicine Concussion is prevalent in both civilian and military (ACRM) defines concussion as having “any alteration of life. Sports-related concussions (SRCs) and recreation- mental state at the time of the accident (dazed, disoriented, related concussions are increasing epidemiological con- or confused),” and the WHO Task Force defines the men- cerns, with annual estimates increasing from 300,000 in 2001 to nearly 4 million in 2011.15,26 Some of this in- crease is probably due to the increased awareness of the condition. American football players are thought to be at an especially high risk of sustaining concussions.7,26,39,61 Concussion and blast-related TBI account for a signifi- cant number of injuries in the military, with incidences in deployed soldiers ranging from 14.9% to approximately 20%, and up to 22%.52,60 It is difficult to identify and di- agnose concussion because this injury can be asympto- matic in its lower limits, and there are no structural ab- normalities present on conventional neuroimaging.3,13,25 Measuring concussion severity in a patient is also chal- lenging because diagnoses and treatments may be based on subjective clinical complaints. Objective assessments for concussion are numerous, but no single test has been validated as the gold-standard diagnostic.16 One interna- tionally accepted method of clinically measuring the se- verity of concussion is the Sport Concussion Assessment FIG. 4. An ACH worn by US Army Spc. Melissa McIntyre of the 793rd 42,97 Military Police Battalion, 8th Military Police Brigade, in Basra, Iraq; Tool, 3rd edition (SCAT3) test. These difficulties are January 20, 2009. Photography by US Armed Forces. Previous source: problematic at the public health and epidemiological level US Department of Defense. Public domain. Figure is available in color because concussion is often underdiagnosed and can eas- online only. ily be overlooked.88,91

J Neurosurg Volume 126 • March 2017 771

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC J. Y. Sone et al.

FIG. 5. Bicycle helmets worn by Olympic cyclists during the 2012 London Olympic Games, July 29, 2012. Photography by Doug Shaw, CC BY-SA 2.0 (https://creativecommons.org/licenses/by-sa/2.0/). Figure is available in color online only.

Methods sion through in simulacra and in silico impact tests.5,47,57,68, 69,83,85,87,89,98 A literature search on PubMed was conducted using These studies are summarized in Table 1. Two of the 10 studies observed military helmet efficacy against key phrases such as “helmet protection against concus- concussion and blast injury in the ACH. Using an in silico sion” or “helmet efficacy in traumatic brain injury” within parametric study, Sarvghad-Moghaddam et al. found that the years 2010–2015 to identify all recent relevant articles. the effectiveness of the ACH against explosive blasts was Reference lists of relevant publications and literature re- direction-dependent.98 Although the study did not specifi- views were searched and used to find additional sources. cally discuss concussive injuries, it showed that fully pro- Studies that were not directly related to finding helmet ef- tected and helmeted systems were most effective against fectiveness in protecting against concussion were elimi- front and top blasts but resulted in adverse effects from nated from the PubMed searches. bottom blasts, by measuring intracranial pressure (ICP) changes, shear stress, and principal strain on the brainstem. Results Unhelmeted systems had more adverse effects from front, The PubMed search terms “helmet” and “concussion” back, and top blasts, but less adverse effects from bottom in all fields yielded 137 results, from which 125 were ex- blasts than fully protected and helmeted systems. Nyein et al. investigated the stress intensity of blasts by measur- cluded and 12 qualified for the literature review. Of the 12 ing the pressure, associated with hydrostatic or volumetric publications under the search terms “helmet” and “concus- tissue deformation, and the equivalent stress, which is as- sion,” 5 articles were impact or in silico laboratory tests, 3 sociated with isochoric distortions in the tissue, through in were neurocognitive evaluations, and 4 were clinical stud- silico blast simulations.83 The study revealed that the exist- ies based on incidence reports. The PubMed search terms ing model of the ACH does not significantly contribute to “helmet” and “head injury” in all fields initially resulted either mitigating or worsening blast effects to the head. in 640 publications, from which 8 articles were selected. Of the 10 publications, 8 articles examined the efficacy Of the 8 studies, 3 were impact or in silico tests, and 5 of civilian helmets against concussion and head injury; 2 were clinical incidence report studies. The PubMed search of those studies examined football helmets’ effectiveness terms “concussion” and “incidence” in the title provided against concussion. Lloyd and Conidi conducted 330 drop 12 results, and 1 incidence report study was selected. One tests on 10 popular football helmets at repeated impacts of article qualified for the review under the search terms 12 mi/hr by using sensor-installed crash test dummy head “eye tracking” and “concussion.” Some publications were and neck systems.57 The study found that football helmets found under multiple sets of search terms. Three of the reduced the risk of skull fracture by 60%–70%, and the 25 publications were located outside of PubMed searches. risk of focal brain contusion by 70%–80%, but the risk of concussion was only reduced by 20%. The study also Impact Simulation of Concussion showed that football helmets were effective against linear Ten studies examined helmet efficacy against concus- acceleration impacts, but not rotational acceleration im-

772 J Neurosurg Volume 126 • March 2017

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC Literature review of helmet efficacy against head injury

TABLE 1. Impact testing and in silico studies observing helmet efficacy against concussion and TBI Authors & Year Method Study Helmet Type No.* Results McIntosh et al., 2013 Impacts ERM Bicycle 66 Unhelmeted subjects had concussion at minimal impact; helmeted subjects had con- cussion at ≥1-m drop, 0–25 km/hr horizontal impacts Rousseau et al., 2014 Impacts ERM Ice hockey 15 Ice hockey helmets were ineffective at reducing concussion risk below 50% for puck impacts at velocity of ≥23 m/sec Lloyd & Conidi, 2014 Impacts ERM Football 330 Football helmets reduced concussion risk by only 20% and were ineffective against rotational acceleration impacts Bartsch et al., 2012 Impacts ERM Football 195 Leather football helmets were comparable or better in head-impact doses and TBI risks than modern helmets Post et al., 2015 Impacts ERM Baseball 84 There was a high concussion risk for subjects regardless of which MLB helmet model was used for protection Karton et al., 2014 Impacts ERM Multiple 84 Speed skating, bicycle, and ice hockey helmets had ≥50% risk of concussion in rota- tional acceleration impacts O’Sullivan et al., 2013 Impacts ERM Taekwondo 32 Headgear not protective against low-energy (50g) and high-energy (150g) impacts that can induce concussion McIntosh & Lai, 2013 Impacts ERM Motorcycle 52 Unhelmeted subjects had concussion at minimal impact; helmeted subjects had con- cussion at ≥1-m drop, ≥25 km/hr horizontal impacts Nyein et al., 2010 In silico ERM ACH NA In silico ACH has no significant effect on mitigating or worsening blast effects to head Sarvghad-Moghad- In silico ERM ACH NA In silico ACH fully protected against front and top blasts and was more effective than no dam et al., 2015 helmet; bottom blasts showed less adverse effect with no helmets ERM = experimental repeated-measures study; MLB = Major League Baseball; NA = not applicable. * Throughout the tables, the heading “No.” refers to values that are either patient sample sizes or number of drops, depending on the study design. pacts. Another study comparing head injury risks associ- at impacts equivalent to a drop height at or above 1 m, and ated with 20th- and 21st-century football helmets through horizontal velocity of 0–25 km/hr for bicycle helmets, and impact tests at 2.0, 3.5, and 5.0 m/sec also showed that at least 25 km/hr for motorcycle helmets. leather 20th-century helmets presented near- and subcon- cussive impact doses and risks similar to or better than Neurocognitive Evaluations on Concussion several 21st-century helmets in multiple impacts in terms Two studies in the neurosurgical literature and 1 study of linear and angular acceleration, angular velocity, the in the sports medicine literature examined neurocognitive Gadd Severity Index (GSI), diffuse axonal injury (DAI), symptoms and scores in helmeted and unhelmeted patients acute SDH, and brain contusion.5 with concussions.58,77,116 Studies on neurocognitive evalu- Four of the 8 studies on civilian helmet efficacy ex- ation are summarized in Table 2. In 1 retrospective study amined the protective efficacy of other sports helmets. of 40 patients with postconcussion syndrome, participants Ice hockey, baseball, and speed skating helmets were all were matched by age and sex to patients with SRC, and shown to have significant risks of concussion by impact neurocognitive evaluations of both acute and subacute velocities or acceleration directions.87,89 Karton et al. found phases were completed using the Post-Concussion Symp- that bicycle, ice hockey, and speed skating helmets all con- tom Scale.77 The study showed that no bivariate associa- sistently presented a greater than 50% probability of play- tions were found between acute-phase and subacute-phase ers wearing these helmets sustaining a concussion at the Post-Concussion Symptom Scale scores and postconcus- mean peak rotational accelerations.47 From striking-pen- sion syndrome development, and that helmet use, loss of dulum impact tests, O’Sullivan et al. showed that standard consciousness (LOC), amnesia, over-the-counter or narcot- Taekwondo helmets do not protect against low-energy (50- ic pain reliever use, and initial consultation with a health g) and high-energy (150-g) head impacts that can induce care provider did not predict postconcussion syndrome. concussion.85 In traffic injury literature, 2 studies from A retrospective study of motocross riders (n = 139) McIntosh and associates tested the efficacy of motorcycle found that 136 (98%) reported “always” wearing a hel- and bicycle helmets.68,69 In the motorcycle helmet test, 52 met while riding, 67 (48%) reported experiencing at least impact tests on 4 helmets were conducted at conditions 1 concussion symptom from riding or racing, 52 of those of 0.5-, 1.0-, and 1.5-m drop heights; 0, 25, and 35 km/hr 67 riders (78%) sought treatment from a health care pro- horizontal striker plate speeds; occipital, lateral, and visor fessional for their symptoms, and 46 of 139 riders (33%) impact orientations; and loose, “2-fingers tight” chin strap reported multiple episodes of concussion symptoms from adjustments in 10 front center impacts.68 In the bicycle hel- any source during the season.58 The study concluded that met test, drop heights of 0.5, 1.0, and 1.5 m, horizontal although almost all riders wore helmets, it was not enough striker plate speeds of 0, 15, and 25 km/hr, and occipital to prevent concussion symptoms after collisions. and lateral head impact orientations were assessed with A third retrospective cohort study matched age, sex, and without helmets.69 Both studies found that unhelmeted number of previous concussions, and days to postconcus- subjects were prone to concussion in the least severe im- sion test to examine acute neurocognitive symptom chang- pact, and helmeted subjects presented risk of concussion es and outcomes after SRC in 138 helmeted and unhel-

J Neurosurg Volume 126 • March 2017 773

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC J. Y. Sone et al.

TABLE 2. Neurocognitive symptoms studies evaluating helmet sults are interpreted with caution because variability in efficacy against concussion and TBI comparison groups was impossible to control and helmet Authors Helmet quality and mode for use was similarly heterogeneous. De- & Year Method Study Type No. Results spite these limitations, the study shows that helmets are not fully protective against concussion as quantitated with Morgan Neuro RCCS Multiple 40 Helmet use, amnesia, SCAT3 or eye tracking. et al., LOC, and over-the- 2015 counter or narcotic Concussion and Head Injury Incidence Studies analgesic use did not There is significant accumulating evidence through predict postconcussive studies on incidences of TBI and concussion that helmets symptoms do not provide the necessary protection against concus- Zucker- Neuro RCS Multiple 138 No differences between sion or even sometimes more severe forms of TBI. Eleven man helmeted and unhelmet- articles examined helmet efficacy through injury reports et al., ed groups in ImPACT of head injury and concussion.6,10–12,21,30,56,66,90,92,111 The in- 2015 scores after head injury cidence and eye movement–tracking studies are summa- Luo et Neuro RCS Moto- 139 Motocross helmet use was rized in Table 3. Of the 11 articles, 3 studies retrospec- al., cross not enough to prevent tively examined ski or snowboard helmet efficacy based 2015 concussion symptoms on patient diagnosis and treatment reports.6,11,92 Baschera after collisions involving et al. showed through logistic regression analysis of 245 head injuries subjects that helmeted and unhelmeted alpine skiers had no significant differences in TBI with respect to helmet Neuro = neurological evaluation; RCCS = retrospective case-control study; use, and although there was an increase in helmet use RCS = retrospective cohort study. from 2000–2001 to 2010–2011, there was no decrease in TBI cases.6 Another study of 678 youth skiers and snowboarders with documented helmet status presented meted athletes by using Immediate Post-Concussion As- 116 through a 95% CI that concussion risk was not associated sessment and Cognitive Testing (ImPACT). With group with helmet use.11 Although Rughani et al. did not specifi- mean differences and reliable change index (RCI) scores cally study helmet use with concussion risks, their study set at the 80% CI, the study found no significant differ- showed that in a retrospective data set of 57 skiers and ences between helmeted and unhelmeted groups for all 4 snowboarders, there was a correlation between reduced neurocognitive tests, and 1 total symptom score (TSS) of skull fractures and helmet use but no significant difference acute, postconcussive outcomes. in the incidence of epidural hematoma and SDH, intrapa- renchymal hemorrhage, subarachnoid hemorrhage (SAH), Eye Movement Tracking or contusion between helmeted and unhelmeted patients.92 Structural damages and the pathophysiology resulting Two studies on amateur boxing presented significant from blast injury or concussion disrupt pathways that con- evidence recommending against headgear use. One arti- trol eye movements mediated by the third and sixth cranial cle observed that mandatory headgear use was associated nerves.94,97 It is well accepted that TBI is associated with with a decrease in referee-stop-contest (RSC) injury (stop- disconjugate eye movements, and abnormal ocular motil- ping the match due to a nonhead injury) and increases in ity is evident in up to 90% of patients with concussion or RSC head injury (stopping the match due to a head injury) blast injury.97,108 Recently our laboratory demonstrated that and RSC injuries overall, ultimately suggesting implemen- the extent of ocular motility dysfunction detected with an tation of the traditional no-headgear rule.12 Preliminary, automated eye tracking algorithm correlates with the se- unpublished research sponsored by the medical commis- verity of symptoms associated with concussion and struc- sion of the International Boxing Association—which was tural TBI.96,97 originally named the Association Internationale de Boxe In a prospective observational study of 55 patients Amateur (AIBA)—suggested that removal of headgear with TBI in helmeted (n = 18) and unhelmeted (n = 37) was associated with reduced concussion rates in amateur groups, Samadani et al. showed that helmeted and unhel- boxers. That is, 7352 rounds with headgear use presented meted subjects did not have statistically significant dif- a concussion rate of 0.38%, whereas 7545 rounds without ferences in 34 eye tracking metrics, ICH evident on CT headgear use resulted in a lower concussion rate of 0.17% scans, hospital admission rate, length of stay (LOS), and (Butler et al., unpublished data, 2013). SCAT3 component scores.95 Forty-two of 89 eye-tracking Of the 11 publications based on injury reports, 2 stud- metrics presented significant differences between nonin- ies evaluated bicycle helmet use regarding concussion and jured control subjects and TBI patient groups at baseline. TBI rates. Castle et al. studied the effects of the mandatory Seven of 89 metrics were significantly worse in unhelmet- helmet law in California starting in 1994 by observing ed patients, and 1 of 89 metrics were worse in helmeted age, sex, race, GCS score, Injury Severity Score, presence patients. Fifteen metrics and 4 SCAT3 component scores of TBI, presence of abdominal injury, and helmet use.21 in the helmeted group and 25 eye-tracking metrics and 3 The study found that there were no significant changes in SCAT3 component scores in the unhelmeted group serial- the rate of helmet use and the rate of TBI in 1684 bike- ly improved from baseline, at 2 weeks, and at greater than related traumas with helmet use between 1992 and 2009. 1-month follow-up. Both groups showed improvement in Although there were more unhelmeted subjects with TBI SCAT3 and eye tracking during recovery. The study’s re- than helmeted subjects, the difference was not significant.

774 J Neurosurg Volume 126 • March 2017

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC Literature review of helmet efficacy against head injury

TABLE 3. Eye movement tracking and incidence report studies on helmet efficacy against concussion and TBI Helmet Authors & Year Method Study Type No. Results Samadani et al., Eye tracking PCSS Multiple 55 Helmeted subjects had relative deficits in eye-tracking metrics and SCAT3 201595 scores; there were no significant differences in CT scan results, admission rate, LOS, or SCAT3 scores between helmeted and unhelmeted groups Baschera et al., Incidence RCS Ski 245 From seasons 2000–2001 to 2010–2011, helmet use increased from 0% to 2015 71.4%, but there were no significant differences in TBI rate, severity, and treatment methods between helmeted and unhelmeted patients Rughani et al., Incidence RCS Ski, snow- 57 Helmet use had more favorable trend in hemorrhage, hematoma, contusion, 2011 board cervical spine injury, need for neurosurgical intervention, and LOS, but was not statistically significant Butler et al., 2013 Incidence Prelim Boxing 14,897 7352 rounds with headgear had 0.38% concussion rate; 7545 rounds without (unpublished) data headgear had 0.17% concussion rate Bergmann et al., Incidence RCCS Ski, snow- 678 LOS, admission, and discharge rates were similar between helmeted and un- 2016 board helmeted groups; unhelmeted subjects were more likely for ICU admission, but it was not statistically significant Bergenstal et al., Incidence RCS Bicycle NA Helmeted concussion rate was 19.4%, and unhelmeted rate was 37.4%, but it 2012 was not significant; the helmeted patients had fewer skull fractures, ICHs Castle et al., 2012 Incidence RCS Bicycle 1684 Implementation of a mandatory helmet law had no significant changes in rates of TBI and helmet use Bianco et al., 2013 Incidence RCS Boxing 29,357 Headgear use was associated with increased RSC and RSC head injury, and decreased RSC injury; study recommended no headgear use Lincoln et al., 2011 Incidence DES Multiple 2651 Helmeted sports had a higher increase than, and 2 times the SRC rate, as unhelmeted sports McGuine et al., Incidence PCS Football 2081 There were no significant differences in SRC rate between Riddell, Schutt, 2014 Xenith helmets Daniels et al., 2015 Incidence RCS Motocross 248 High TBI rates after pediatric motocross/motorbike accidents were observed despite predominant helmet use by riders Rowson et al., Incidence RCS Football 1833 Revolution helmet model had a significantly lower SRC rate than VSR4 model, 2014 at 2.82% vs 4.47%, respectively DES = descriptive epidemiological study; PCS = prospective cohort study; PCSS = prospective case-series study; Prelim = preliminary.

Another retrospective cohort study on pediatric bicyclists with 2081 athletes and 134,437 football exposures, 206 reported a concussion rate of 19.4% in helmeted patients players (9%) sustained 211 SRCs (1.56/1000 exposures).66 and a concussion rate of 37.4% in unhelmeted patients, but That study showed that no significant difference in SRC the difference was not statistically significant between the incidence existed between players wearing Riddell (1171 2 patient groups (p = 0.0509).10 The reduced incidences helmets, 9.1% incidence [95% CI 7.6%–11.0%]); Schutt of skull fractures and ICH in helmeted pediatric subjects (680 helmets, 8.7% incidence [95% CI 6.7%–11.1%]); or Xe- were statistically significant (p = 0.0408, p = 0.0079). nith (436 helmets, 9.2% incidence [95% CI 6.7%–12.4%]) The literature on other sports and recreational hel- type helmets. However, another retrospective analysis in met efficacy based on concussion incidences consistently the neurosurgical literature showed conflicting results. showed that helmets did not fully protect against concus- Rowson et al. examined head impacts and concussion in- sion and its risks. A retrospective population-based cohort cidences of 1833 collegiate football players who wore ei- study of 60 cases of TBI in pediatric motocross accidents ther a Riddell VSR4 or a Riddell Revolution helmet, using observed a significantly high TBI incidence (57 cases with a 95% CI.90 In their study, 3.34% of all players sustained LOC, pathological findings on CT scans in 10 patients) concussions, 4.47% of players in VSR4 helmets sustained despite helmet use (43 of 60 cases, 71.6%).30 One descrip- concussions, and 2.82% of players in Revolution helmets tive epidemiology study observing SRC in 12 different sustained concussions. The difference in concussion inci- high school sports with 2651 concussion incidences from dences between VSR4 and Revolution helmets was found 1997–1998 to 2007–2008 showed in a 95% CI that hel- to be statistically significant (c2 = 4.68, p = 0.0305). meted sports had a higher increase in and nearly twice the concussion rate as in unhelmeted sports.56 Such studies are limited by bias in that helmeted sports are more likely to Discussion require contact and collisions. In this review of helmet efficacy against TBI and con- Two studies of football helmets examined protective ef- cussion, we report that helmets do not fully protect against ficacy between different types of helmets but not between concussion. The majority of studies showed that helmet use helmeted and unhelmeted players.66,90 In 1 cohort study did not result in a statistically significant reduction in con-

J Neurosurg Volume 126 • March 2017 775

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC J. Y. Sone et al. cussion incidence and symptoms. Wide varieties of helmets the introduction to our review, the WHO and the ACRM were unable to protect against the impacts of simulated have discrepancies in their definition of concussion.88,91 directional forces that would induce concussion in human Furthermore, the interchangeability between the defini- subjects. Patients who wore helmets at the time of head tions of mTBI and concussion are still debated, although injury accidents and collisions presented concussive and there is a trend to distinguish them as separate intracra- postconcussive symptoms in neurological examinations af- nial injuries.13,33,64,65,100,110 This lack of standardization cre- ter the injury. Although eye movement tracking has shown ates inconsistent clinical comparisons between different its effectiveness in quantitatively measuring the severities studies, inhibiting a clear consensus on the injury and its of concussion and TBI, eye-tracking metrics did not pro- outcomes. 2) There is also a lack of standardization in ex- vide significant differences between helmeted and unhel- amining helmet efficacy. This also stems from the lack of meted patients admitted to the hospital. In addition, TBI a standardized definition for concussion, which creates and concussion incidence reports presented that high rates inconsistent comparisons of the helmets’ efficacy against of helmet use were not associated with low concussion in- the specific head injury. Additionally, although most stud- cidence and favorable changes in concussion incidence fol- ies focus on cranial pathology, some studies will include lowing the implementation of mandatory helmet use regu- patients sustaining neck injury, vestibular dysfunction, or lations. Some studies indicated efficacy and a favorable other noncranial pathology. Although these inclusions do trend of helmet use in reducing TBI as an aggregate group not necessarily detract from the strength of the evidence or in reducing TBI with pathological findings on CT scans, on helmet efficacy, they can shift the focus of helmet ef- such as traumatic SAH or SDH. However, there were no ficacy away from concussion and other head injuries. articles reporting that helmeted patients presented statis- The quality of the acquired data in the studies is often tically significantly reduced rates of concussion or better a major limitation to the examination of helmet efficacy clinical outcomes of concussion than unhelmeted patients. against concussion and TBI. Several studies examined by Earlier literature reviews have presented similar findings this review were limited by their study designs as retro- on helmet efficacy against TBI and concussion; that is to spective, self-reporting, or nonrandomized in helmet use. say, helmets do not provide the necessary protection from Some studies also presented some selection bias in acquir- such traumatic intracranial injuries.9,14,28,67 ing data on helmeted and unhelmeted patients. It is dif- ficult to overcome the selection bias for acute concussive Older Literature on Helmet Efficacy injuries and especially more severe forms of TBI for sev- Studies published between 2000 and 2009 also present- eral reasons. 1) Acute head injuries are typically examined ed consistent evidence that helmets are limited or ineffec- by the on-site or nearest health care provider to ensure the tive in their protection against concussion. Three studies on patient’s safety and the data’s accuracy. Prospective stud- rugby players in the sports medicine literature showed that ies involving SRC and helmet efficacy typically use on-site the use of headgear did not reduce the rates of concussion health care providers who assess the concussive symp- or head injury, or result in any statistically significant dif- toms.56,66,90 This is especially true for more severe forms ferences between helmeted and unhelmeted groups.62,70,71 of TBI because such injuries entail high risks of morbidity A cluster randomized controlled trial with a control group and mortality, and it is essential for the patient to receive (n = 1493), a standard headgear group (n = 1128), and a immediate care. These necessities limit the injured sub- modified headgear group (n = 1474) found that incidence ject groups to smaller, sometimes underpowered cohorts rate ratios for the standard headgear group referenced to specific to the sport and region. 2) Helmets and their effi- controls were 0.95 and 1.02 for game and missed-game cacy are also specific to the occupation or sport, requiring injuries, whereas ratios for the standard headgear group separate subgroups and preventing randomization across referenced to the unhelmeted group were 1.11 and 1.10.71 the different helmet types. These limitations decrease the A prospective cohort study on rugby players in Dunedin, statistical power and favor smaller, localized cohorts in the New Zealand, found no evidence of a statistically signifi- studies. cant efficacy of helmet use against concussion in a 95% The inability to perform truly blinded, randomized CI.62 A pilot study involving 15 rugby unions showed simi- prospective trials of helmeted versus unhelmeted subjects lar evidence that the use of headgear in rugby was not asso- due to ethical considerations and the knowledge that hel- ciated with statistically significant improvement in protec- mets prevent TBI-induced death is another critical limi- tion against concussion compared with the players who did tation in studies examining helmet efficacy. The inability not wear helmets.70 However, 1 study on the use of helmets to conduct a blinded study also stems from the fact that in football (soccer) in the sports medicine literature sug- the health care provider for the TBI needs to know about gested that use of protective headgear was associated with helmet use for proper neurological assessment. It is also a reduction in the risk of head injuries such as concussion.31 difficult to acquire TBI subjects for the unhelmeted group because many occupations and sports require helmet use Limitations to Helmet Efficacy in Studies and Practical as part of safety regulations. Furthermore, animal model Aspects studies cannot supplant human subjects for prospective There are several major limitations to the current stud- blinded randomized trials because 1) the helmet designs ies and their experimental designs regarding helmet effi- worn by humans may not necessarily be reproducible or cacy, which in turn impose limitations on their literature scalable at a rodent model scale, and 2) the neurological reviews. These limitations are also shared by any literature evaluations of rodent models are not directly comparable reviews on concussion. 1) There is a definite lack of stan- to human subject evaluations. dardization for a definition of concussion. As delineated in The efficacy of helmets against concussion and TBI

776 J Neurosurg Volume 126 • March 2017

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC Literature review of helmet efficacy against head injury faces many limitations beyond studies, experimental de- and TBI may also lead to a reduced neurosurgical burden sign, and literature reviews. Bias in personality and risk- regarding preventable deaths and perioperative complica- taking behavior can result in preferences toward or against tions due to the pathophysiology of TBI and concussion. helmet use. The bias in personality also affects the likeli- Examination of studies on helmet efficacy within the neu- hood that the patient will seek a physician after the po- rosurgical literature is summarized in Table 4. tential head injury and ensure sufficient postinjury care. Beyond the redesign of helmets, additional studies The quality, condition, and proper fit of the helmet used by should be conducted to further examine helmet efficacy an individual cannot necessarily be controlled to optimize against concussion with simulated impact tests and assess- helmet efficacy, even though such factors are crucial for ment of physiological function, with eye tracking, neuro- protection against concussion and TBI. It is expected that cognitive outcomes, and epidemiological studies. To im- researchers fail to demonstrate efficacy against concussion prove the understanding of whether helmets are effective (which is difficult to quantitate) because it is already diffi- against concussion, these studies would ideally emphasize cult to demonstrate efficacy against the need for neurosur- larger sample sizes of differently helmeted subjects and gical intervention, due to the fact that most studies are sig- would incorporate better study designs that are prospec- nificantly underpowered and difficult to balance because tive, randomized, and controlled for confounding factors. of the limitations described in this section. As described previously, ethical considerations place sig- nificant limitations on prospective, randomized, and con- Current Implications of Helmet Efficacy trolled study designs for examining helmet efficacy. It may Evidence continues to indicate that helmets are signifi- be possible to conduct prospective, randomized controlled cantly protective against death due to brain injury, more trials with helmeted and unhelmeted groups in sports that severe forms of TBI, skull fractures, and traumatic ICHs. incorporate optional helmet use, such as rugby union, Therefore helmet use is recommended for sports, activi- without violating ethical concerns. However, randomiza- ties, and occupations that entail risks of intracranial inju- tion may be difficult, and selection bias may thus still ex- ry. Helmets may be less effective against concussion than ist. Prospective randomized trials may also be possible against more severe forms of TBI due to the inherently with different brands of helmets. This type of study would different biomechanics of these injuries. 1) Concussion improve helmet efficacy by examining which helmet has is induced by low-energy linear and rotational accelera- greater protection and recommending the safer helmet to tions that cause more diffuse, distributed impact loading the general population. The disadvantages of such a study and peak ICP than the brain tissue can tolerate, resulting design are that 1) there are no unhelmeted subjects, and in damage to the microscopic structures of the brain and 2) a substantially large sample size of TBI is needed to temporary onset of neurological impairment.2,75,101 2) More ensure equal and statistically significant data for different severe forms of TBI are typically caused by high-energy helmets. focal forces loading onto a localized region of the brain, Diagnostic instruments used to measure concussion resulting in injuries such as penetrating TBI and depressed should be improved to increase helmet efficacy. More skull fractures.2,75 3) Closed types of severe TBI are in- robust diagnosis and definition of concussion can lead to duced by distributive impacts that result in macroscopic better helmet standards based on a refined understanding lesions and inflammation, such as DAI.55 It is more dif- of concussion biomechanics and pathophysiology. In addi- ficult to design helmets and component materials that can tion, patients who have experienced a concussion are more safely reduce the low-energy directional accelerations and likely to suffer recurrent concussions, and patients with a peak pressures that are distributed throughout the brain chronic history of concussions have higher risks of devel- than to make helmets that can absorb severe diffuse im- oping neurodegenerative diseases such as chronic trau- pacts or stop externally localized forces strong enough to matic encephalopathy (CTE) and dementia.32,43,86,105 Thus, penetrate the dura mater or damage the skull. reforming concussion diagnostics may result in not only an increase in helmet efficacy but also a decrease in epi- Implications for Helmet Efficacy in the Future demiological severity of recurrent concussions and neuro- Modern helmets for civilian and military purposes have degenerative diseases through earlier and more compre- begun to focus more on protection against concussion and hensive clinical management of postconcussive symptoms blast injuries. This review highlights the fact that the lack and stronger precautions to minimize future concussive of helmet efficacy against such intracranial trauma is a injuries. Expanding the awareness of the general popula- growing medical and public health concern due to the epi- tion with regard to TBI and concussion is an additional demic prevalence of TBI and concussion and their increas- method to reduce such injuries, by increasing individual ingly recognized morbidity and mortality.27,80 Because the precaution in activities that entail risk of concussion and most effective method of protection against concussion TBI and pressing for improvements in safety regulations and TBI is prevention, reconsideration of helmet design in sports and transportation. and efficacy is crucial to reducing incidences of concus- sion, TBI, secondary neuronal damage, and chronic neu- A Role for Physicians and Engineers in Improving Helmet rocognitive disability at clinical and epidemiological lev- Efficacy els. The importance of materials engineering for helmets It is incumbent on neurosurgeons, neurologists, and should be further promoted and emphasized, because new other neuroscientists to provide leadership and expertise synthetic materials comprising the helmets are critical to for the improvement of helmet efficacy against concus- reducing the acceleration and shock of concussive impacts. sion and TBI. Clinicians and scientists should collaborate Improved helmet efficacy and design against concussion with engineers in designing helmets that not only protect

J Neurosurg Volume 126 • March 2017 777

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC J. Y. Sone et al.

TABLE 4. Neurosurgical literature examining helmet efficacy against concussion and head injury Authors & Year Method Study Helmet No. Results Bartsch et Impacts ERM Football 195 Leather football helmets were comparable or better in head impact doses and TBI risks than al., 2012 modern helmets Morgan et Neuro RCCS Multiple 40 Helmet use, amnesia, LOC, and over-the-counter or narcotic analgesic use did not predict al., 2015 postconcussive symptoms Luo et al., Neuro RCS Motocross 139 Motocross helmet use was not enough to prevent concussion symptoms after collisions 2015 involving head injuries Rughani et Incidence RCS Ski, snow- 57 Helmet use had a more favorable trend in hemorrhage, hematoma, contusion, cervical spine al., 2011 board injury, need for neurosurgical intervention, and LOS, but was not statistically significant Daniels et Incidence RCS Motocross 248 High TBI rates after pediatric motocross/motorbike accidents were observed despite predomi- al., 2015 nant helmet use by riders Rowson et Incidence RCS Football 1833 Revolution helmet model had significantly lower SRC rate than VSR4 model, at 2.82% vs al., 2014 4.47%, respectively against impact injuries, but also minimize the risks of DAI T-tau levels were also shown to be acutely higher after a due to acceleration and deceleration of the head. For ex- single incidence of concussion in ice hockey players, in ample, it is important to reduce the mass of the helmet which the elevated concentrations 1 hour after the injury because heavier helmets can potentially cause greater were associated with the time needed to resolve concus- magnitudes of injury through greater inertia. Consider that sive symptoms and the time needed for recovery before flexion-extension of the neck during trauma may actually returning to play.99 Standardizing the pathology of struc- be exacerbated if the mass of the head is greater, result- tural TBI and concussion with quantitative assessments is ing in greater shear injury within the brain. Engineers and the crucial foundation for allowing studies to have stronger clinicians need to work together to create helmets that not evidence, more clearly defined injury groups, and be di- only protect against focal injury resulting in death or skull rectly comparable with one another. Standardization also fracture, but also reduce shear injury. facilitates care for concussion and TBI, reducing overall There is also a need for greater focus on the develop- clinical burden on not only neurosurgery but also other ment of objective, standardized, quantitative diagnostic relevant specialties. assessments for concussion, TBI, and other head injuries within the neurosurgical specialty. Currently, there are nu- Conclusions merous concussion and TBI assessment techniques under investigation, such as quantitative electroencephalography, There is significant evidence to support helmet use to eye tracking, and tau protein level changes in CSF and protect against devastating intracranial injury. A consid- blood.64,101 Preliminary data and several studies suggest erable amount of evidence indicates that contemporary that quantitative electroencephalography measures signifi- helmets are not efficacious in protecting against concus- cantly different brain activity in patients after acute SRC, sion. Further research and development on helmet design chronic recovery of SRC, PCS, and posttraumatic migraine and efficacy is needed to reduce concussion in helmeted following concussion relative to control subjects.4,48,49,51,63 subjects. Ocular motility dysfunction induced by structural TBI and concussion was able to be detected by an automated eye- References tracking algorithm, in which the metrics positively corre- 1. Agarwal V, Tisherman S: Traumatic brain injury, in Falter lated with the horizontal disconjugacy of eye movements F, Screaton NJ (eds): Imaging the ICU Patient. Springer: and the severity of the injury’s symptoms.96,97 Twenty- London, 2014, pp 365–375 five measures from the eye-tracking algorithm were also 2. Bandak FA, Ling G, Bandak A, De Lanerolle NC: Injury shown to be significantly different in patients with concus- biomechanics, neuropathology, and simplified physics of sion compared with control subjects.97 explosive blast and impact mild traumatic brain injury. Handb Clin Neurol 127:89–104, 2015 Recent studies have shown the potential of tau protein 3. Barr WB: Mild traumatic brain injury, in Sherer M, as a viable biomarker for structural TBI, mTBI, and con- Sander AM (eds): Handbook on the Neuropsychology of cussion. Tau is a structural microtubule-associated protein Traumatic Brain Injury. New York: Springer, 2014, pp expressed in neurons, and is especially abundant in neuro- 347–369 nal axons.76,115 The hyperphosphorylation of tau (P-tau) can 4. Barr WB, Prichep LS, Chabot R, Powell MR, McCrea M: induce the destabilization of axonal microtubules and the Measuring brain electrical activity to track recovery from formation of toxic neurofibrillary tangles, which are found sport-related concussion. Brain Inj 26:58–66, 2012 in the brains of people with neurodegeneration such as Al- 5. Bartsch A, Benzel E, Miele V, Prakash V: Impact test com- 73,74,76,115 parisons of 20th and 21st century American football hel- zheimer disease and CTE. Total tau (T-tau) in the mets. J Neurosurg 116:222–233, 2012 CSF and blood plasma was found to be present in signifi- 6. Baschera D, Hasler RM, Taugwalder D, Exadaktylos A, cantly higher concentrations in boxers who experienced Raabe A: Association between head injury and helmet use acute repeated mTBI than in control subjects, in whom the in alpine skiers: cohort study from a Swiss level 1 trauma higher T-tau levels normalized after 2 weeks.81,82 Plasma center. J Neurotrauma 32:557–562, 2015

778 J Neurosurg Volume 126 • March 2017

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC Literature review of helmet efficacy against head injury

7. Baugh CM, Kiernan PT, Kroshus E, Daneshvar DH, 28. Daneshvar DH, Baugh CM, Nowinski CJ, McKee AC, Montenigro PH, McKee AC, et al: Frequency of head- Stern RA, Cantu RC: Helmets and mouth guards: the role of impact-related outcomes by position in NCAA Division I personal equipment in preventing sport-related concussions. collegiate football players. J Neurotrauma 32:314–326, Clin Sports Med 30:145–163, x, 2011 2015 29. Daneshvar DH, Riley DO, Nowinski CJ, McKee AC, Stern 8. Benson BW, Hamilton GM, Meeuwisse WH, McCrory P, RA, Cantu RC: Long-term consequences: effects on normal Dvorak J: Is protective equipment useful in preventing con- development profile after concussion. Phys Med Rehabil cussion? A systematic review of the literature. Br J Sports Clin N Am 22:683–700, ix, 2011 Med 43 (Suppl 1):i56–i67, 2009 30. Daniels DJ, Clarke MJ, Puffer R, Luo TD, McIntosh AL, 9. Benson BW, McIntosh AS, Maddocks D, Herring SA, Wetjen NM: High occurrence of head and spine injuries in Raftery M, Dvorák J: What are the most effective risk- the pediatric population following motocross accidents. J reduction strategies in sport concussion? Br J Sports Med Neurosurg Pediatr 15:261–265, 2015 47:321–326, 2013 31. Delaney JS, Al-Kashmiri A, Drummond R, Correa JA: The 10. Bergenstal J, Davis SM, Sikora R, Paulson D, Whiteman C: effect of protective headgear on head injuries and concus- Pediatric bicycle injury prevention and the effect of helmet sions in adolescent football (soccer) players. Br J Sports use: the West Virginia experience. W V Med J 108:78–81, Med 42:110–115, 2008 2012 (Abstract) 32. Delaney JS, Lacroix VJ, Gagne C, Antoniou J: Concussions 11. Bergmann KR, Flood A, Kreykes NS, Kharbanda AB: among university football and soccer players: a pilot study. Concussion among youth skiers and snowboarders: a review Clin J Sport Med 11:234–240, 2001 of the National Trauma Data Bank From 2009 to 2010. 33. Dimou S, Lagopoulos J: Toward objective markers of con- Pediatr Emerg Care 32:9–13, 2016 cussion in sport: a review of white matter and neurometa- 12. Bianco M, Loosemore M, Daniele G, Palmieri V, Faina M, bolic changes in the brain after sports-related concussion. J Zeppilli P: Amateur boxing in the last 59 years. Impact of Neurotrauma 31:413–424, 2014 rules changes on the type of verdicts recorded and implica- 34. Dow RS, Ulett G, Raaf J: Electroencephalographic studies tions on boxers’ health. Br J Sports Med 47:452–457, in head injuries. J Neurosurg 2:154–169, 1945 2013 35. Erichsen JE: On Concussion of the Spine, Nervous Shock 13. Blennow K, Hardy J, Zetterberg H: The neuropathology and and Other Obscure Injuries of the Nervous System: neurobiology of traumatic brain injury. Neuron 76:886– In Their Clinical and Medico-legal Aspects. London: 899, 2012 Longmans, Green, and Company, 1875 14. Bonfield CM, Shin SS, Kanter AS: Helmets, head injury 36. Foulis J: Reports and analyses and descriptions of new and concussion in sport. Phys Sportsmed 43:236–246, inventions, in medicine, surgery, dietetics, and the allied 2015 sciences. BMJ 1:24–25, 1889 15. Broglio SP, Eckner JT, Martini D, Sosnoff JJ, Kutcher JS, 37. Galarneau MR, Woodruff SI, Dye JL, Mohrle CR, Wade Randolph C: Cumulative head impact burden in high school AL: Traumatic brain injury during Operation Iraqi Freedom: football. J Neurotrauma 28:2069–2078, 2011 findings from the United States Navy-Marine Corps Combat 16. Broglio SP, Ferrara MS, Macciocchi SN, Baumgartner TA, Trauma Registry. J Neurosurg 108:950–957, 2008 Elliott R: Test-retest reliability of computerized concussion 38. Gerberding JL, Binder S: Report to Congress on Mild assessment programs. J Athl Train 42:509–514, 2007 Traumatic Brain Injury in the United States: Steps 17. Cairns H: Head injuries in motor-cyclists. The importance to Prevent a Serious Public Health Problem. Atlanta: of the crash helmet. BMJ 2:465–471, 1941 National Center for Injury Prevention and Control, Centers 18. Cairns H: Treatment of head injuries in war. BMJ 1:1029– for Disease Control and Prevention, 2003 1030, 1940 39. Gessel LM, Fields SK, Collins CL, Dick RW, Comstock 19. Cairns H, Holbourn H: Head injuries in motor-cyclists: with RD: Concussions among United States high school and col- special reference to crash helmets. BMJ 1:591–598, 1943 legiate athletes. J Athl Train 42:495–503, 2007 20. Carey ME, Sacco W, Merkler J: An analysis of fatal and 40. Grujicic M, Bell WC, Pandurangan B, Glomski PS: Fluid/ non-fatal head wounds incurred during combat in Vietnam structure interaction computational investigation of blast- by U.S. forces. Acta Chir Scand Suppl 508:351–356, 1982 wave mitigation efficacy of the advanced combat helmet. J 21. Castle SL, Burke RV, Arbogast H, Upperman JS: Bicycle Mater Eng Perform 20:877–893, 2011 helmet legislation and injury patterns in trauma patients 41. Haring RS, Narang K, Canner JK, Asemota AO, George under age 18. J Surg Res 173:327–331, 2012 BP, Selvarajah S, et al: Traumatic brain injury in the elder- 22. Chiu WT, Chu SF, Chang CK, Lui TN, Chiang YH: ly: morbidity and mortality trends and risk factors. J Surg Implementation of a motorcycle helmet law in Taiwan and Res 195:1–9, 2015 traffic deaths over 18 years. JAMA 306:267–268, 2011 42. Hoshizaki TB, Post A, Oeur RA, Brien SE: Current and 23. Chiu WT, Huang SJ, Tsai SH, Lin JW, Tsai MD, Lin TJ, future concepts in helmet and sports injury prevention. et al: The impact of time, legislation, and geography on the Neurosurgery 75 (Suppl 4):S136–S148, 2014 epidemiology of traumatic brain injury. J Clin Neurosci 43. Iverson GL, Gardner AJ, McCrory P, Zafonte R, Castellani 14:930–935, 2007 RJ: A critical review of chronic traumatic encephalopathy. 24. Chiu WT, Kuo CY, Hung CC, Chen M: The effect of the Neurosci Biobehav Rev 56:276–293, 2015 Taiwan motorcycle helmet use law on head injuries. Am J 44. Javouhey E, Guérin AC, Chiron M: Incidence and risk fac- Public Health 90:793–796, 2000 tors of severe traumatic brain injury resulting from road 25. Corso P, Finkelstein E, Miller T, Fiebelkorn I, Zaloshnja E: accidents: a population-based study. Accid Anal Prev Incidence and lifetime costs of injuries in the United States. 38:225–233, 2006 Inj Prev 12:212–218, 2006 45. Jefferson G: War wounds and air raid casualties. War 26. Covassin T, Swanik CB, Sachs ML: Epidemiological con- wounds of the head: I. BMJ 2:347–349, 1939 siderations of concussions among intercollegiate athletes. 46. Jeffreys J: The British Army in India. London: Longman, Appl Neuropsychol 10:12–22, 2003 Brown, Green, Longmans, & Roberts, 1858 27. Dacey RG Jr, Alves WM, Rimel RW, Winn HR, Jane JA: 47. Karton C, Rousseau P, Vassilyadi M, Hoshizaki TB: The Neurosurgical complications after apparently minor head evaluation of speed skating helmet performance through injury. Assessment of risk in a series of 610 patients. J peak linear and rotational accelerations. Br J Sports Med Neurosurg 65:203–210, 1986 48:46–50, 2014

J Neurosurg Volume 126 • March 2017 779

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC J. Y. Sone et al.

48. Kontos AP, Reches A, Elbin RJ, Dickman D, Laufer I, incidence of sport-related concussion in high school football Geva AB, et al: Preliminary evidence of reduced brain players: a multifactorial prospective study. Am J Sports network activation in patients with post-traumatic migraine Med 42:2470–2478, 2014 following concussion. Brain Imaging Behav [epub ahead 67. McIntosh AS, Andersen TE, Bahr R, Greenwald R, Kleiven of print], 2015 S, Turner M, et al: Sports helmets now and in the future. Br 49. Korn A, Golan H, Melamed I, Pascual-Marqui R, Friedman J Sports Med 45:1258–1265, 2011 A: Focal cortical dysfunction and blood-brain barrier dis- 68. McIntosh AS, Lai A: Motorcycle helmets: head and neck ruption in patients with Postconcussion syndrome. J Clin dynamics in helmeted and unhelmeted oblique impacts. Neurophysiol 22:1–9, 2005 Traffic Inj Prev 14:835–844, 2013 50. Kulkarni SG, Gao XL, Horner SE, Zheng JQ, David NV: 69. McIntosh AS, Lai A, Schilter E: Bicycle helmets: head Ballistic helmets—their design, materials, and performance impact dynamics in helmeted and unhelmeted oblique against traumatic brain injury. Compos Struct 101:313– impact tests. Traffic Inj Prev 14:501–508, 2013 331, 2013 70. McIntosh AS, McCrory P: Effectiveness of headgear in a 51. Kutcher JS, McCrory P, Davis G, Ptito A, Meeuwisse WH, pilot study of under 15 rugby union football. Br J Sports Broglio SP: What evidence exists for new strategies or tech- Med 35:167–169, 2001 nologies in the diagnosis of sports concussion and assess- 71. McIntosh AS, McCrory P, Finch CF, Best JP, Chalmers ment of recovery? Br J Sports Med 47:299–303, 2013 DJ, Wolfe R: Does padded headgear prevent head injury in 52. Laker SR: Epidemiology of concussion and mild traumatic rugby union football? Med Sci Sports Exerc 41:306–313, brain injury. PM R 3 (10 Suppl 2):S354–S358, 2011 2009 53. Langlois JA, Rutland-Brown W, Thomas KE: Traumatic 72. McIntosh AS, Patton DA, Fréchède B, Pierré PA, Ferry E, Brain Injury in the United States: Emergency Barthels T: The biomechanics of concussion in unhelmeted Department Visits, Hospitalizations, and Deaths. Atlanta: football players in Australia: a case-control study. BMJ Centers for Disease Control and Prevention, National Center Open 4:e005078, 2014 for Injury Prevention and Control, 2004 73. McKee AC, Robinson ME: Military-related traumatic brain 54. Lidell JA: On fractures of the skull, restricted to the inner injury and neurodegeneration. Alzheimers Dement 10 (3 table. Am J Med Sci 83:334–335, 1882 Suppl):S242–S253, 2014 55. Lin Y, Wen L: Inflammatory response following diffuse 74. McKee AC, Stern RA, Nowinski CJ, Stein TD, Alvarez VE, axonal injury. Int J Med Sci 10:515–521, 2013 Daneshvar DH, et al: The spectrum of disease in chronic 56. Lincoln AE, Caswell SV, Almquist JL, Dunn RE, Norris traumatic encephalopathy. Brain 136:43–64, 2013 JB, Hinton RY: Trends in concussion incidence in high 75. Meaney DF, Smith DH: Biomechanics of concussion. Clin school sports: a prospective 11-year study. Am J Sports Sports Med 30:19–31, vii, 2011 Med 39:958–963, 2011 76. Medina M, Avila J: New perspectives on the role of tau in 57. Lloyd J, Conidi F: Comparison of common football helmets Alzheimer’s disease. Implications for therapy. Biochem in preventing concussion, hemorrhage and skull fracture Pharmacol 88:540–547, 2014 using a modified drop test. (P5.320). Neurology 82:P5.320, 77. Morgan CD, Zuckerman SL, Lee YM, King L, Beaird S, 2014 (Abstract) Sills AK, et al: Predictors of postconcussion syndrome after 58. Luo TD, Clarke MJ, Zimmerman AK, Quinn M, Daniels sports-related concussion in young athletes: a matched case- DJ, McIntosh AL: Concussion symptoms in youth control study. J Neurosurg Pediatr 15:589–598, 2015 motocross riders: a prospective, observational study. J 78. Moss WC, King MJ, Blackman EG: Skull flexure from Neurosurg Pediatr 15:255–260, 2015 blast waves: a mechanism for brain injury with implications 59. Maas AI, Stocchetti N, Bullock R: Moderate and severe for helmet design. Phys Rev Lett 103:108702, 2009 traumatic brain injury in adults. Lancet Neurol 7:728–741, 79. Murphy TE, Baker DI, Leo-Summers LS, Tinetti ME: 2008 Trends in fall-related traumatic brain injury among older 60. Mac Donald CL, Adam OR, Johnson AM, Nelson EC, persons in Connecticut from 2000–2007. J Gerontol Werner NJ, Rivet DJ, et al: Acute post-traumatic stress Geriatr Res 3:3, 2014 symptoms and age predict outcome in military blast concus- 80. Narotam PK, Morrison JF, Nathoo N: Brain tissue oxygen sion. Brain 138:1314–1326, 2015 monitoring in traumatic brain injury and major trauma: out- 61. Marar M, McIlvain NM, Fields SK, Comstock RD: come analysis of a brain tissue oxygen-directed therapy. J Epidemiology of concussions among United States high Neurosurg 111:672–682, 2009 school athletes in 20 sports. Am J Sports Med 40:747–755, 81. Neselius S, Brisby H, Theodorsson A, Blennow K, 2012 Zetterberg H, Marcusson J: CSF-biomarkers in Olympic 62. Marshall SW, Loomis DP, Waller AE, Chalmers DJ, Bird boxing: diagnosis and effects of repetitive head trauma. YN, Quarrie KL, et al: Evaluation of protective equipment PLoS One 7:e33606, 2012 for prevention of injuries in rugby union. Int J Epidemiol 82. Neselius S, Zetterberg H, Blennow K, Randall J, Wilson 34:113–118, 2005 D, Marcusson J, et al: Olympic boxing is associated with 63. McCrea M, Prichep L, Powell MR, Chabot R, Barr WB: elevated levels of the neuronal protein tau in plasma. Brain Acute effects and recovery after sport-related concussion: Inj 27:425–433, 2013 a neurocognitive and quantitative brain electrical activity 83. Nyein MK, Jason AM, Yu L, Pita CM, Joannopoulos JD, study. J Head Trauma Rehabil 25:283–292, 2010 Moore DF, et al: In silico investigation of intracranial blast 64. McCrory P, Meeuwisse WH, Aubry M, Cantu B, Dvorák J, mitigation with relevance to military traumatic brain injury. Echemendia RJ, et al: Consensus statement on concussion Proc Natl Acad Sci U S A 107:20703–20708, 2010 in sport: the 4th International Conference on Concussion 84. Okie S: Traumatic brain injury in the war zone. N Engl J in Sport held in Zurich, November 2012. Br J Sports Med Med 352:2043–2047, 2005 47:250–258, 2013 85. O’Sullivan DM, Fife GP, Pieter W, Shin I: Safety perfor- 65. McCrory P, Meeuwisse WH, Echemendia RJ, Iverson GL, mance evaluation of taekwondo headgear. Br J Sports Med Dvorák J, Kutcher JS: What is the lowest threshold to make 47:447–451, 2013 a diagnosis of concussion? Br J Sports Med 47:268–271, 86. Petraglia AL, Dashnaw ML, Turner RC, Bailes JE: Models 2013 of mild traumatic brain injury: translation of physiological 66. McGuine TA, Hetzel S, McCrea M, Brooks MA: Protective and anatomic injury. Neurosurgery 75 (Suppl 4):S34–S49, equipment and player characteristics associated with the 2014

780 J Neurosurg Volume 126 • March 2017

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC Literature review of helmet efficacy against head injury

87. Post A, Karton C, Hoshizaki TB, Gilchrist MD, Bailes J: in children and youths: a review of research since 1990. J Evaluation of the protective capacity of baseball helmets for Child Neurol 31:20–27, 2016 concussive impacts. Comput Methods Biomech Biomed 105. Trotter BB, Robinson ME, Milberg WP, McGlinchey RE, Engin 19:366–375, 2015 Salat DH: Military blast exposure, ageing and white matter 88. Roozenbeek B, Maas AI, Menon DK: Changing patterns in integrity. Brain 138:2278–2292, 2015 the epidemiology of traumatic brain injury. Nat Rev Neurol 106. Ulfarsson T, Lundgren-Nilsson Å, Blomstrand C, Jakobsson 9:231–236, 2013 KE, Odén A, Nilsson M, et al: Ten-year mortality after 89. Rousseau P, Hoshizaki TB, Gilchrist MD: For ASTM F-08: severe traumatic brain injury in western Sweden: a case Protective capacity of ice hockey player helmets against control study. Brain Inj 28:1675–1681, 2014 puck impacts, in Ashare A, Ziejewski M (eds): Mechanism 107. Vaaramo K, Puljula J, Tetri S, Juvela S, Hillbom M: Head of Concussion in Sports. West Conshohocken, PA: ASTM trauma with or without mild brain injury increases the risk International, 2014, pp 196–207 of future traumatic death: A controlled prospective 15-year 90. Rowson S, Duma SM, Greenwald RM, Beckwith JG, Chu follow-up study. J Neurotrauma 32:1579–1583, 2015 JJ, Guskiewicz KM, et al: Can helmet design reduce the risk 108. Ventura RE, Jancuska JM, Balcer LJ, Galetta SL: of concussion in football? J Neurosurg 120:919–922, 2014 Diagnostic tests for concussion: is vision part of the puzzle? 91. Ruff RM, Iverson GL, Barth JT, Bush SS, Broshek DK: J Neuroophthalmol 35:73–81, 2015 Recommendations for diagnosing a mild traumatic brain 109. Viano DC, Halstead D: Change in size and impact perfor- injury: a National Academy of Neuropsychology education mance of football helmets from the 1970s to 2010. Ann paper. Arch Clin Neuropsychol 24:3–10, 2009 Biomed Eng 40:175–184, 2012 92. Rughani AI, Lin CT, Ares WJ, Cushing DA, Horgan MA, 110. Walker WC, Cifu DX, Hudak AM, Goldberg G, Kunz RD, Tranmer BI, et al: Helmet use and reduction in skull frac- Sima AP: Structured interview for mild traumatic brain tures in skiers and snowboarders admitted to the hospital. J injury after military blast: inter-rater agreement and devel- Neurosurg Pediatr 7:268–271, 2011 opment of diagnostic algorithm. J Neurotrauma 32:464– 93. Rustemeyer J, Kranz V, Bremerich A: Injuries in combat 473, 2015 from 1982–2005 with particular reference to those to the 111. Wang SS: Boxing group bans headgear to reduce concus- head and neck: A review. Br J Oral Maxillofac Surg sions. Wall Street Journal. March 14, 2013. (http://www. 45:556–560, 2007 wsj.com/articles/SB10001424127887323393304578360250 94. Samadani U, Farooq S, Ritlop R, Warren F, Reyes M, 659207918) [Accessed March 18, 2016] Lamm E, et al: Detection of third and sixth cranial nerve 112. Wasserman RC, Waller JA, Monty MJ, Emery AB, palsies with a novel method for eye tracking while watching Robinson DR: Bicyclists, helmets and head injuries: a rider- a short film clip. J Neurosurg 122:707–720, 2015 based study of helmet use and effectiveness. Am J Public 95. Samadani U, Kim A, Kolecki R, Reyes M, Ritlop R, Wall Health 78:1220–1221, 1988 S, et al: Helmets may provide partial protection against con- 113. Woolley LC: Excavations at Ur. London: Ernest Benn cussion: a prospective study with SCAT3 and eye tracking. Limited, 1954 J Neurotrauma 32:A-31, 2015 (Abstract) 114. Xydakis MS, Fravell MD, Nasser KE, Casler JD: 96. Samadani U, Li M, Qian M, Laska E, Ritlop R, Kolecki R, Analysis of battlefield head and neck injuries in Iraq and et al: Sensitivity and specificity of an eye movement track- Afghanistan. Otolaryngol Head Neck Surg 133:497–504, ing based biomarker for concussion. Concussion [epub 2005 ahead of print], 2015 115. Zhang Z, Song M, Liu X, Kang SS, Kwon IS, Duong DM, 97. Samadani U, Ritlop R, Reyes M, Nehrbass E, Li M, Lamm et al: Cleavage of tau by asparagine endopeptidase mediates E, et al: Eye tracking detects disconjugate eye movements the neurofibrillary pathology in Alzheimer’s disease. Nat associated with structural traumatic brain injury and concus- Med 20:1254–1262, 2014 sion. J Neurotrauma 32:548–556, 2015 116. Zuckerman SL, Lee YM, Odom MJ, Forbes JA, Solomon 98. Sarvghad-Moghaddam H, Jazi MS, Rezaei A, Karami G, GS, Sills AK: Sports-related concussion in helmeted vs. Ziejewski M: Examination of the protective roles of hel- unhelmeted athletes: who fares worse? Int J Sports Med met/faceshield and directionality for human head under 36:419–425, 2015 blast waves. Comput Methods Biomech Biomed Engin 18:1846–1855, 2015 99. Shahim P, Tegner Y, Wilson DH, Randall J, Skillbäck T, Pazooki D, et al: Blood biomarkers for brain injury in Disclosures concussed professional ice hockey players. JAMA Neurol Uzma Samadani has equity in Oculogica, Inc., an eye-tracking 71:684–692, 2014 neurodiagnostic company. 100. Signoretti S, Lazzarino G, Tavazzi B, Vagnozzi R: The pathophysiology of concussion. PM R 3 (10 Suppl Author Contributions 2):S359–S368, 2011 101. Stone JL, Patel V, Bailes JE: The history of neurosurgical Conception and design: Sone, Samadani. Acquisition of data: treatment of sports concussion. Neurosurgery 75 (Suppl Sone. Analysis and interpretation of data: Sone, Samadani. 4):S3–S23, 2014 Drafting the article: Sone. Critically revising the article: all 102. Suderman BL, Hoover RW, Ching RP, Scher IS: The effect authors. Reviewed submitted version of manuscript: all authors. of hardhats on head and neck response to vertical impacts Approved the final version of the manuscript on behalf of all from large construction objects. Accid Anal Prev 73:116– authors: Sone. 124, 2014 103. Thompson DC, Rivara FP, Thompson R: Helmets for pre- Correspondence venting head and facial injuries in bicyclists. Cochrane Je Yeong Sone, Department of Neurosurgery, New York Database Syst Rev (2):CD001855, 2000 University School of Medicine, 145 E 32nd St., 5th Fl., New 104. Thurman DJ: The epidemiology of traumatic brain injury York, NY 10016. email: [email protected].

J Neurosurg Volume 126 • March 2017 781

Unauthenticated | Downloaded 10/05/21 04:27 PM UTC