Clinical manifestations

Presentation and course

A wide range of sporting and recreation-related injuries may occur of the peripheral and central nervous systems. The clinical manifestation depends on the sport or recreation involved, the specific nature of the acute injury, the nature of a repetitive motion leading to a chronic injury, as well as the age of the participant. Specific clinical manifestations are described below within each individual sport considered.

Prognosis and complications

Prognosis varies greatly depending on the nature of the injury in the athlete and is not specific to any one sport. Bruce and Echemendia performed both traditional and computerized neuropsychiatric testing on college athletes with a remote history of concussion and found no patterns of impairment compared to controls (57). Slobounev and colleagues performed multimodal testing on collegiate athletes after concussion and found visual-motor dysfunction that persisted after the postconcussive symptoms lapsed. They also found markedly reduced rates of recovery in patients with multiple concussions (415).

Alves reported a landmark 4-year study of concussions in college football players who were all tested pre-season and then after concussions, with comparisons made to controls with solely orthopedic injuries. The study documented both the presence and duration of subjective complaints (headache, dizziness, and memory disturbance) as well as presence and duration of impairments on a simple neuropsychological battery. The time frame of the subjective and objective impairments coincided, with attention and concentration deficiencies being of greatest duration (08).

Lovell and colleagues examined high school athletes after concussion and then again after clinical recovery with a battery that included subjective scales, neurocognitive testing, and functional MRI evaluation (244). Abnormalities on functional MRI within a week of the injury predicted a longer duration of symptoms and objective impairments on testing.

McAllister and associates examined 214 college football and ice hockey players over the course of their seasons and compared them to athletes in noncontact sports. They found that 24% of contact sport athletes had poorer than predicted outcomes on post-season new learning tasks, compared to 4% in controls. Poor performance correlated to head injury exposure (258).

It has been theorized that premorbid psychological conditions may potentiate the effects of closed head injury and lengthen recovery time. Nelson and associates evaluated 127 high school athletes suffering concussions and found that both severity of concussion as well as preinjury somatic symptom score both influenced recovery time (301). Merritt and Arnett came to a similar conclusion linking pre-injury affective disorder to recovery in a college injury cohort (274). Goreth and Palokas examined this in a Cochrane-type meta-analysis of 12 studies involving nearly 3000 injuries (147). Trends were seen demonstrating prolonged recovery time in pediatric patients with pre-injury learning disabilities, anxiety, depression, poor academic performance, somatization, and sleep disorders, although the heterogeneity of endpoints precluded definitive conclusions.

Clinical vignette

A 24-year-old male presented to clinic with a 2-week history of bilateral hand weakness and numbness over the fourth and fifth digits bilaterally. The day of his current symptomatology origin was spent snowmobiling. During that day, he was bothered by excessive vibration of the handlebars and the discomfort this created over his hands. He obtained duct tape and had his friend duct tape his hands to the handlebars in an attempt to dampen the bothersome vibration. This allowed him to continue snowmobiling the remainder of the day. That night, he began to notice tingling over his fourth and fifth digits on the volar aspects bilaterally, and by the next morning, he noted numbness over this same distribution. Over the next few days, he noted weakness of the hands when attempting to hold objects and perform pinching actions. Examination demonstrated bilateral loss of pinprick sensation over the volar portions of digits 4 and 5 bilaterally, as well as bilateral severe weakness of the dorsal interossei muscles and abductor digiti quinti minimi (ADQM) bilaterally, with other intrinsic muscles of the hands demonstrating normal strength. Tinel sign was negative at the cubital tunnel but was positive at the wrist at Guyon canal. Froment sign was present bilaterally. Nerve conduction studies demonstrated a markedly increased distal motor latency and reduction in compound motor action potential of the ulnar nerve conduction to the first dorsal interossei and ADQM bilaterally. The sensory nerve action potential at the fifth digit with stimulation of the ulnar nerve at the wrist demonstrated prolonged latency and small amplitude bilaterally as well. More proximal ulnar nerve conduction studies around the elbow were unremarkable. The patient was diagnosed with bilateral ulnar neuropathy at Guyon Canal affecting the hypothenar, deep palmar, and sensory branches of the ulnar nerve. Recommendations were made to avoid the offensive activity, and the patient’s symptoms resolved over the subsequent 4 to 5 months.

Biological basis

Etiology and pathogenesis

Auto racing. At the Indianapolis Raceway Park, a retrospective study over six seasons identified neurologic injuries to drivers during 61 open-wheel racing events, with four drivers requiring admission to hospital and two to intensive care for head injuries (423). Head injuries account for 29% of all injuries in professional auto racing, with open head injuries composing only 5% of these head injuries (137; 449).

High vehicle g-forces of over 50 g are more likely to lead to head injury (473). Closed head injuries have rarely included intracranial hemorrhages, and more commonly included diffuse axonal injury (137). Exposure to emissions of carbon monoxide and vehicle fires was studied in race car drivers using a breath analyzer; an increase in carboxyhemoglobin concentrations during the competition event was identified in all race car drivers although no correlation with carbon monoxide level and driver symptomatology was demonstrated (175). Heat stroke has rarely been associated with auto racing (192). Spinal injuries compose 20% of injuries suffered by professional auto drivers, which most commonly occur with a vehicular rollover and lead to cervical spine or spinal cord injury (449). The brachial plexus is at risk of injury due to the tight fastening of arm to helmet to prevent centrifugal force for auto drivers, and the sciatic and peroneal nerves are subject to compression due to the small size of the car cockpit (449).

Ballet or professional dancing. Painless, isolated weakness of external rotation of the right arm of a professional dancer with clinical and electrophysiological evidence of suprascapular neuropathy has been reported. The injury was possibly secondary to repetitive, forceful arm movements with external rotation and abduction and postulated entrapment of the nerve at the spinoglenoid notch. Nearly complete recovery of muscle function occurred after 4 months of rest from dance (95; 226; 483). Femoral neuropathy has been reported in dancers who perform repeated simultaneous hip extension and knee flexion (the “Horton Hinge”) (276; 373) and has been associated with patellar dislocation in a jazz dancer (402). Peroneal and sural neuropathy due to tight ribbons and elastics in dancing shoes has also been reported (374). Dorsal cutaneous neuritis has been apparent in dancers who sit on their feet with pressure on the dorsum of the foot (374). Just as in active runners, serious dancers can occasionally present with a Morton neuroma of the plantar nerves (468).

The rare syndrome of fibrocartilaginous embolism of the intervertebral disc leading to spinal cord infarction has occurred in a previously healthy 30-year-old ballet dancer after an intensive training session (425).

Baseball. Injuries in baseball revolve around the use of a high-speed projectile (baseball), a repetitive activity (throwing the baseball, swinging the bat), and acrobatic maneuvers (catching the baseball). The most frequent mechanism of baseball-related injury is being hit by the ball, which represents 62% of acute injuries (324). In particular, local nerve lesions are common in baseball players. Many of the peripheral nerve injuries occur with the act of pitching. Suprascapular nerve injuries presenting as shoulder pain accompanied by weakness of shoulder abduction and external rotation can occur in a pitching arm (359; 88). Entrapment of the suprascapular nerve in pitchers may occur at the suprascapular or spinoglenoid notches (241). Often, suprascapular nerve injury can mimic a rotator cuff tear. In particular, starting pitchers (not relief pitchers) and more experienced pitchers are at greatest risk for infraspinatus muscle atrophy (89). Position players are far less likely to develop a suspected suprascapular nerve palsy (89). Windmill pitching, used in softball, has been associated with radial neuropathy at different anatomical sites (412). The "thrower's fracture" is a distal humeral fracture leading to radial nerve palsy identified in softball pitchers (90). Ulnar neuropathy at the elbow is common among baseball pitchers (97; 165; 489; 242; 13). In one study of 72 professional baseball players undergoing arthroscopic or open elbow surgery, ulnar neuropathy was diagnosed in 15% (13). Adolescent players have developed cubital tunnel syndrome, presenting initially as medial elbow pain. Surgical treatment with anterior subcutaneous transposition of the ulnar nerve relieved symptoms for each of six players for up to 3.3 postoperative years. At surgery, medial protrusion of the triceps irritating the ulnar nerve or fibrosis surrounding the ulnar nerve was observed (14). Humeral shaft fracture in softball (227) and baseball (310) pitchers have been associated with radial nerve palsies. In adults with humeral fractures associated with throwing a baseball, 16% of patients had concurrent radial nerve palsy (310). Pronator syndrome in the proximal forearm may occur in the pitching arm with entrapment of the median nerve by fibrous bands of the pronator teres (242). Thoracic outlet syndrome has presented as numbness in the fingers of the throwing hand of a college baseball player, with compression of the neurovascular bundle demonstrated using magnetic resonance angiography with arm held in abduction (117). Quadrilateral space syndrome within the pitching arm of a baseball pitcher has been reported in one case with compression of the distal axillary nerve and partial compression of the posterior humeral circumflex artery (82). Batters may be susceptible to a traumatic neuroma of the ulnar digital nerve of the thumb (25).

Injuries due to more acute causes are probably more common than chronic injuries in baseball. In childhood baseball players, concussion comprises about 1% of all injuries (350). Baseball players are also subject to mild traumatic brain injuries at all ages (343). More serious intracranial injuries within the sport of baseball are rare, with one report of epidural hematoma secondary to a baseball bat striking the head (355). Other more serious head injuries have occurred in high school and college players due to getting hit by a pitched ball, collision with another fielder, or collision with a base runner, and some of those injuries result in fatality (41). Pitchers are at risk for being struck by a batted ball, which can lead to serious head injuries that may result in coma or be associated with subdural hematoma (41). Fielders struck by a batted ball in the head have suffered fatal head injuries (41). Cervical fractures have also occurred in high school and college players due to collisions with other players, leading to quadriplegia or death in some cases (41). Little league players may suffer catastrophic injuries due to sliding, having infielders struck by a batted ball, or batters hit by a pitch (286). A retrospective study of baseball bat-related injuries revealed craniocerebral injury being the most frequent form of injury and most frequent cause of death. Of all victims struck in the head, 26% sustained an intracranial hemorrhage (152). Even rarer is the spinal cord injury in baseball players, with a single report of acute central cervical spinal cord syndrome in a 32-year-old male baseball player after hyperextension injury (283).

Basketball. Although basketball injuries are relatively common, neurologic causes of injury are relatively rare in basketball players. A single report of a suprascapular nerve lesion without any history of shoulder girdle trauma has been reported in a basketball player, perhaps due to repeated nerve traction over the coracoid notch during basketball dunking (451). Recovery in this player was nearly complete after 3 weeks of inactivity (451). The burner, or stinger, described in football players has rarely been seen in basketball players (123). Other forms of neuropathy have not been reported in traditional basketball players, but compression neuropathies of the arms are common injuries in wheelchair basketball players, particularly carpal tunnel syndrome (188).

Mild traumatic brain injuries occur with low incidence in high school basketball players and are due to striking the basketball pole or rim or being struck by a falling pole or backboard (73).

Bicycling. Recreational and competitive cycling, including BMX biking, road cycling, and off-road biking, is associated with a wide range of neurologic injuries affecting the peripheral and central nervous system. Probably the most common form of nerve entrapment is that of the ulnar nerve at the wrist, seen among bicyclists and resulting in weakness of grip and numbness of the fourth and fifth digits (242). In cyclists, the most common location of ulnar nerve entrapment is within Guyon Canal (181). Within a large cross-sectional study of 160 male professional cyclists, 30% reported paraesthesia or numbness in the fingers, mostly from the ulnar innervated region (11). Besides the ulnar nerve, the median nerve may also be abnormal in many cyclists. In one study of professional cyclists, symptoms of carpal tunnel syndrome occurred in 25% of hands, with 62% of symptomatic hands tested demonstrating abnormal electrodiagnostic findings on stimulation of the median nerve (69). Bilateral median nerve compression may also occur in the cyclist (51).

A unique form of neuropathy in the cyclist is the pudendal neuropathy. In fact, “bicycle seat neuropathy” is one of the most common injuries reported by cyclists (478). One report of male competitive cyclists documented symptoms of recurrent numbness of penis and scrotum after prolonged cycling, along with an altered sensation of ejaculation and micturition and reduced awareness of defecation. Cyclists may develop pudendal neuropathies secondary to racing-bicycle saddles applying pressure on the perineum. Changes in bike saddle position and riding technique led to symptomatic improvement (406). In a larger cross-sectional study of 160 male professional cyclists, 22% reported symptoms of penile numbness or hypesthesia after a long duration of cycling, and 13% reported transient impotence for weeks. Eighty-five percent of cyclists reporting genital numbness and impotence also reported numbness within the hands after cycling, perhaps suggesting liability (11). In another study of cyclists participating in a 500-mile bicycle tour, 45% of cyclists reported mild or transient perineal numbness, 10% reported severe symptomatology, and 2% reported temporary breaks in riding due to symptoms (477). Not exclusive to male cyclists, 34% of female cyclists also reported perineal numbness (231). Bicycle seat neuropathy may be due to entrapment of the pudendal nerve passing through the Alcock canal enclosed by ischial bone and obturator internus (308). Adjustment of the bike seat by tilting the nose of the seat down and lowering the bike seat position to relieve pressure on the perineum may be beneficial (308). Another neuropathy due to prolonged bicycle use is posterior cutaneous thigh neuropathy (264). A unique injury dubbed “pedal pusher’s palsy” with bilateral sciatic nerve palsies presented following prolonged unicycle riding (143).

Head injuries due to bicycle riding have become a hot political topic, with many cities in Canada and selected districts within the United States, leading to the institution of mandatory bicycle helmet laws. In one Canadian study, helmet use was associated with less likelihood of hospital admission following injury, a lower incidence of head and facial injury, and a lower incidence of concussion (239). In Canadian recreational bicyclists, isolated crashes are the most common cause of injury, although collision with vehicles accounts for 64% of bicycle-related deaths (218). An American study also found that the use of bicycle safety helmets for children led to a lower incidence of skull fractures and may have prevented deaths due to head injuries that occurred in helmet-less children (393). Rarely, falling on bicycle handles has been associated with severe head injuries and even death (04). Bicycle safety helmet legislation in California has been associated with statistically significant reductions in head injuries among bicyclists aged 17 years and under, although there was no statistically significant change in injury outcomes for adult bicyclists (234). For children beginning cycling, injury rates are reduced if the age of first starting bicycling debut is delayed until 7 or 8 years instead of at 4 or 5 years (167).

Although anecdotal suggestions of bicycle accidents being more common with BMX-style bikes have occurred, the incidence of injuries as compared to non-BMX bikes appears to be similar (491). In fact, BMX riders had a lower proportion of serious injuries than racing bicycle riders, including fewer head injuries. Most accidents occurring in BMX riders are related to performing stunts or poor cycling technique (491). The proportion of concussions as an injury related to BMX riding is approximately 7% (187). Although most injuries occurring in off-road bicycle racing are musculoskeletal, 12% of injuries occur over the cranial region (210), and a small incidence of concussion (less than 1%) occurs, about 40% less than in other forms of cycling (224; 361). This low incidence may relate to a significantly higher helmet usage rate in off-road bikers (361). Interestingly, women appeared to be much more likely than men to sustain a serious injury while off-road biking (224; 225), although males are more frequent participants and, therefore, suffer the bulk of the injuries (210), including cervical spinal cord injuries (16). Bicycling accidents have been associated with cervical spinal cord injury in children (322) and thoracic disc herniation leading to myelopathy in one female adult (495).

Bodybuilding and weightlifting. Bodybuilding can be associated with a range of entrapment neuropathies secondary to repetitive motions or excessive muscle bulk, sometimes associated with anabolic steroid use. Often, the nature of the mononeuropathies in the bodybuilder or weightlifter is unique. Repeated bench presses over 2 weeks in a young male was associated with weakness of interossei, fourth and fifth lumbricals, adductor pollicis, and abductor digiti minimi secondary to injury to the deep motor branch of the ulnar nerve with severe conduction block identified on sequential nerve conduction studies (282). Ulnar neuropathy has also been reported in a 39-year-old competitive male weightlifter due to suspected compression between the heads of flexor carpi ulnaris (93; 378). A professional bodybuilder presenting with proximal forearm pain and supinator tenderness analogous to radial tunnel syndrome secondary to compression of the posterior interosseous nerve was reported. In this case, power squats were possibly implicated, and conservative treatment led to symptomatic improvement (107). Another unique weightlifting-associated neuropathy is a progressive bilateral medial pectoral neuropathy secondary to postulated pectoralis minor hypertrophy and subsequent intramuscular entrapment of the medial pectoral nerves (222; 367). A compressive lesion of the deep palmar branch of the ulnar nerve was reported in a patient who entered an intensive program of push-ups on a hard floor (464). Other forms of mononeuropathy reported in bodybuilders include suprascapular neuropathy (03; 222), carpal tunnel syndrome, ulnar neuropathy at the elbow, long thoracic neuropathy, lateral antebrachial cutaneous neuropathy (222), and musculocutaneous neuropathy (47; 242; 31). Other mononeuropathies reported in bodybuilders using anabolic steroids with clinical and electrophysiological findings include femoral, thoracodorsal, dorsoscapular, and the terminal branch of the suprascapular nerve (151; 133; 385; 31; 280). In most reports, resting of the affected region with entrapment neuropathy has been associated with recovery of strength and muscle bulk. Brachial plexus injuries are rare in the bodybuilder, but “stingers” (defined in Football section) have rarely been reported in weightlifters (123). A single case report of vigorous bodybuilding has been associated with "rectus abdominis syndrome" with elevated creatine kinase and rectus abdominis biopsy demonstrating features of acute rhabdomyolysis (380). Lastly, benign exertional headache presenting as occipital and cervical pain, perhaps in response to a Valsalva maneuver, occurs in weightlifters and may respond to isometric neck-strengthening exercises (357).

Neuropathies in the lower extremities are extremely rare in weightlifters. Bilateral peroneal neuropathy due to an acute exertional compartment syndrome in each leg has occurred in a male concurrently using anabolic steroids (238).

Boxing. Essentially all neurologic injuries resulting from boxing involve intracranial contents as opposed to spinal cord or peripheral nervous system injury, and the sequelae of boxing range from the acute complication of concussion to the chronic dementing process of dementia pugilistica, to death.

The assessment of neurologic injuries due to boxing requires two important considerations: first, acute neurologic injuries should be distinguished from chronic brain injuries; second, the level of competitive boxing, amateur versus professional, must be considered. Amateur boxing differs from professional boxing in terms of the duration of fights, the nature of rules and regulatory policies, the degree of medical evaluation, and the use of protective devices (ie, headgear). Obviously, acute neurologic injuries such as concussion, post-concussion syndrome, and intracranial hemorrhage are easily identified as compared to chronic neurologic injuries because of their immediate impact and obvious relationship to recently inflicted trauma. Serious acute intracranial injuries due to boxing are recognized but felt to be rare (456). In the case of a knockout, the boxer has sustained a concussion, which is easily recognizable. One roadblock limiting the assessment of the incidence of more serious injuries in boxing is the potential minimization of injuries due to regulatory policy. Concussion has gained more recognition as a medical condition requiring attention following trauma (208), but attempts to quantify the number of concussions in boxing have not been published. Computerized cognitive assessment in military cadet boxers has demonstrated inferior performances on simple reaction time and continuous performance tests after a boxing-induced concussion as compared with baseline testing (469). Due to shorter careers and fewer career bouts, as well as improved medical monitoring, today’s boxers will likely be subject to a lower incidence of traumatic brain injuries (78).

The measure of more chronic effects of boxing and neuropsychological and cognitive effects has been studied many times with various results. For example, a study of former Swedish amateur boxers identified only a single task to be performed inferiorly relative to other athletes: finger-tapping (289). A lack of laboratory abnormalities in boxers relative to other athletes, including testing with electroencephalography and brainstem auditory evoked responses, has also been noted (163). Chronic neurologic injuries from boxing tend to have an insidious onset and often present and continue to progress after cessation of boxing (195). Of ex-professional boxers who had participated in the sport for at least 3 years, 17% were found to have clinical evidence of CNS deficit felt to be attributable to boxing (362). These abnormalities included cerebellar, extrapyramidal, and intellectual impairments, and the most severe form of post-boxing encephalopathy has been termed dementia pugilistica. Other clinical features have included tremor, dysarthria, and psychiatric changes, such as explosive behavior and paranoid and jealous delusions (273). In another observational study, risk factors for CNS deficit after boxing included professional boxing as opposed to amateur boxing, number of punches taken, and lack of “scientific ability” (86). In contrast, few amateur boxers demonstrated no abnormalities on detailed neurologic examinations and MRI of the brain (198). Although amateur boxers and controls do not demonstrate abnormalities in regional cerebral blood flow (rCBF) with inhalation of 133-xenon, professional boxers demonstrate diffuse reductions in regional cerebral blood flow, especially in the frontocentral regions (363). Perhaps this reduction in perfusion may relate to chronic brain injury over regions of brain felt to be affected in dementia pugilistica and recurrent concussion (363).

Baird and colleagues determined that between 1950 and 2007, there were an estimated 339 fatalities associated with boxing (21). A significant reduction was noted after 1983, which they speculated was due to a reduction in rounds for championship fights. More fatalities were seen in lower weight classes, and most occurred within the ring.

Cervical spine fractures are rare in boxing, yet potentially catastrophic. A transient spinal cord injury occurred in a young male boxer who had an os odontoideum, which may have placed him at risk for such an injury (340). Another boxer sustained a C6 vertebral body fracture and quadriplegia during a boxing match (209).

Usually seen in football players, the “burner” or “stinger” is rarely reported in boxers (123).

Diving and scuba diving. The predominant mechanism of injury associated with water activities is diving, with cervical spinal cord injury composing 4.9% of all water-related accidents in children (186). Among all sports, diving is the most common cause of spinal cord injury (21.6%) and most commonly affects males (88%) of young age (mean of 28.5 years) (203). Patients sustaining a cervical spine injury are more likely to be in the early-mid teenage years (10 years to 14 years in one study) and to be experienced divers (35; 186). The second most common group to suffer high cervical spinal cord injuries due to diving is young adult male divers (154). The cervical spinal cord injury is so common in divers that in one large retrospective study, all spinal cord injuries associated with diving were at the cervical level (381). The most commonly affected vertebral levels are C5 and C6 (05; 219). Almost all spinal cord injury related to diving (87%) occurs in private or residential swimming pools (103). Alcohol and the presence of a pool party were respectively involved in 49% and 46% of cases of diving-associated spinal cord injury in a retrospective study. As would be expected, most diving-related spinal cord injuries (57%) occur with diving into less than 4 feet of water. The absence of a lifeguard on duty was noted in 94% of cases in a retrospective study. Ordinary diving accounts for 70% of spinal cord injury cases, with unusual or trick dives less commonly associated (103). The fracture type most associated with worsening neurologic outcomes is the teardrop fracture, as compared with the burst fracture (05).

In older adults, rare reports of internal carotid artery occlusion and carotid dissection have occurred following a sports dive (184). Cerebral arterial oxygen gas bubble emboli have also been clinically associated with alterations of consciousness, seizures, and focal neurologic deficits (303). At autopsy, gas emboli can be detected in cerebral and spinal arteries (320). Cerebral infarction has been reported in professional breath-hold divers during repetitive dives, with MRI demonstrating multiple T2-weighted hyperintensities corresponding to their neurologic deficit (214). A relationship between neurologic decompression illness, of both cerebral and spinal dysfunction, and the presence of cardiopulmonary shunt was demonstrated in a blinded, controlled study. Larger shunts that are present without a Valsalva maneuver were significantly more common in divers affected by neurologic decompression illness (485). Paradoxical thromboembolism and neurologic decompression illness in divers have been associated with large sizes of patent foramen ovale, which may be amenable to septal occlusion (467).

Headaches reported in divers include “skip breathing” headache with migraine-like features, possibly associated with carbon dioxide accumulation (15). Another headache presenting with faciotemporal steady discomfort in divers is a headache due to an excessively tight face mask or goggles, perhaps due to pressure on superficial cutaneous nerves or the supraorbital nerve (334).

One form of neuropathy that occurs in scuba divers is lateral femoral cutaneous neuropathy due to compression of the diver’s weight belt on the nerve (264).

Equestrian or horse racing. The union of two species in a sport is uncommon and can lead to injuries due to a lack of “teamwork” as well as high speeds achieved and a long distance for the rider to fall to the ground. Surprisingly, injuries in equestrian sports are common, perhaps 20 times more common than motorcycling (140).

Closed head injuries are a common occurrence in riders due to falls from the horse. In fact, the majority of equestrian-related injuries (60%) are caused by ejection of the rider by the horse or falling from the horse (220), whereas a smaller percentage of injuries (40%) is due to being kicked by a horse, even as a bystander (220). In one Canadian retrospective study, closed head injury was the most common cause of admission to hospital after equestrian-related trauma (420). Closed head injuries of various severities have been associated with a fourth nerve palsy in 46% of patients and loss of vision in 20% (125). Occurrence of intracranial hematoma has been reported with equestrian trauma, including pediatric cases, and may be associated with mortality (257; 220). When compared with other common mechanisms of pediatric injury, the average degree of head injury in riders was only less than in automobile-related accidents (44). Equestrian-related deaths are secondary to head injury in 57% of cases for riders under 25 years old, which also account for most hospitalizations in this population (32; Nelson and 32). Females injured in equestrian injuries outnumber the numbers of males, partly due to female predominance in this activity (33). However, over the age of 44 years, more men are injured than women, and male deaths far outnumber female deaths above the age of 64 years (33). Helmet use reduces the risk and severity of head injuries and should be vigorously promoted (125), as most riders are helmetless (220). Use of a helmet in pediatric riders was associated with decreased frequency and severity of central nervous system injury (44). Although childhood horse riding injuries are not as publicized as adult injuries, their severity appears to be high. In children, the severity of equestrian-related injuries ranks second only to pedestrians being struck by a car, with a higher score than all-terrain vehicle, bicycle, and passenger motor vehicle crash injuries (189).

An entirely different sport from equestrian, professional horse racing is a fast sport with high injury rates (453). There are surprisingly few epidemiological reports on the subject and only one report of cervical spinal cord injuries in two German jockeys (10).

The long-term effects of horse riding on the cervical and lumbar spine of jockeys may be deleterious as well due to repetitive trauma. The incidence of degenerative changes of both cervical and lumbar spine was higher in jockeys versus age-matched controls in a prospective study with both clinical and radiographic evaluation of the spine (450).

Although not as common as head injuries, spinal injuries, including quadriplegic or paraplegic injuries, make up 30% of severe injuries (220). Injuries to celebrities such as Christopher Reeve may lead to greater injury prevention in equestrian riders, including helmet use and education (404).

Peroneal neuropathy has developed secondary to a compartment syndrome following injury due to a fall while recreational horse riding (02).

Football. American and Canadian football is an aggressive sport with significant physical contact and high risk of injury. In most studies, football is reported as the sport most likely to be associated with injury, serious injury, as well as neurologic injury. Even with the presence of helmets, the high velocity and violent nature of football leads to many head injuries. Of 10 sports investigated, football accounted for 63% of mild traumatic brain injuries (343). The presence of a helmet for protection can also lead to injuries, as 7% of all football injuries involve being struck by an opponent's helmet (73). The published incidences of concussion in football have significant variation, partly due to over-reporting of recalled episodes of concussion in teammates when compared with self-reports and videotape analysis (265). Concussions occur in football at an estimated rate of 6.1 per athlete season in one study (77), more than twice the incidence of other team sports. High school football players self-reported an incidence of concussion of 47% over one season, with 35% of all players reporting multiple concussions (229). In high school football players, concussion rates are 33.09 per 100,000 athlete exposures (386). Another study of both high school and college football players reported only 5% of players sustained one concussion, whereas 15% of those players sustained a second concussion during the same season (156). The most common specific diagnosis in Canadian varsity players was concussion (271). A study of Canadian Football League professional players suggested a 45% concussion incidence rate over one season, with a 70% incidence of multiple concussions in players reporting at least one (99). In college football players, the incidence of concussion was equally distributed between games and practice (100), which is unique for this sport and injury type as most sports-related injuries have higher incidences during competitions. A slightly increased incidence of concussion was noted among offensive and defensive linemen (09; 100), as well as special teams players. Blocking may lead to more concussions than tackling (09). Most impacts (71%) leading to concussion are from an opponent’s helmet, arm, or shoulder pad to the side of the player’s helmet, and often these impacts are on the highest portion of the helmet (331). The most common symptoms of mild traumatic brain injury in professional football players were headaches (55.0%), dizziness (41.8%), and blurred vision (16.3%) (326). A loss of consciousness occurs in 8.9% to 9.3% of cases (156; 326). Features that predict a long time (greater than 7 days) for recovery from concussion symptoms include disorientation to time, retrograde amnesia, fatigue, and a loss of consciousness (327). Of those players suffering concussions where prolonged time is required to return to the game, the highest frequency occurs in quarterbacks (14.8%) (329). The signs and symptoms with the highest incidence for leaving the game for greater than 7 days are disorientation to time, retrograde amnesia, and fatigue. Loss of consciousness for more than 1 minute is also a predictor for missing more than 7 days of football (329).

The presence of an episode of concussion may place the player at increased risk for recurrence of up to three to four times (139; 156), which may place the patient at increased risk for second impact syndrome (265). Although publicized in the past, the prevalence of second-impact syndrome is difficult to calculate. The second impact syndrome has been defined as a sustained head injury after an initial head injury, usually a concussion, where symptoms associated with the first injury have not fully cleared. It has been postulated that this second impact leads to the rapid development of cerebral vascular congestion and increased intracranial pressure, resulting in brainstem herniation and death (265). Although some of the first reported patients with second impact syndrome were football players (also ice hockey and boxing), skepticism about the true nature of this entity continues. Despite this, greater education is obviously required as one study suggested that only 18% to 23% of concussed college and professional football players realized they had suffered a concussion (99; 100). The most vulnerable professional football players to suffer repeated concussions are the ball return carriers on special teams and quarterbacks (330). More than 70% of Canadian university football players reported symptoms of concussion over 1 year of play in a retrospective study, and 85% of these players reported more than one episode of concussion (100). Fortunately, the majority of concussions are of grade I (88% to 95%) (09; 229). As has been suspected by many professional players, the use of artificial turf may play a role in the high incidence of concussion in football players (297). However, the presence of a custom-made mouthguard does not seem to influence the incidence of concussion in collegiate level football players (486). The short-term effects of concussion include neurocognitive impairment when high school and college football players were assessed immediately and 15 minutes after injury as compared to a preseason baseline score (262). Players with a concussion and loss of consciousness (grade 3 concussion) were found to be the most severely impaired immediately after injury, giving a gradation for the effects of concussion on neurocognition (262). College football players with prior concussions demonstrate learning disability for the Trail-Making Test and Symbol Digit Modalities Test with neuropsychological performance testing as compared to players without prior concussions recorded (81). Video analysis of professional football plays resulting in head injuries and laboratory reconstructions using helmeted dummies have determined that concussions are primarily related to translational acceleration at the time of impact as well as a significant velocity change (331). Head-to-head collision has been simulated via a finite element head model, which suggests that shear stress at the brainstem may be an injury predictor for concussion (500). A head-down stance appears to increase the risk of concussion in professional football players (462). As well, players who tackle with the head down and use the head as a battering ram may be at increased risk for more severe forms of injuries to the head and neck (65).

The presence of posttraumatic headaches in football players after hits in the game is common (21%), particularly among offensive and defensive linemen and defensive backs, and is often unreported to trainers (372). The most vulnerable players for losing more than 7 days with concussion are quarterbacks and the defensive secondary (327). Although controversial, players with milder grades of concussions without evidence of general symptoms, somatic complaints, cranial nerve effects, cognition problems, memory problems, or unconsciousness do not appear to be at greater risk for subsequent head injury in the same game or season (328).

More severe brain injuries are less commonly reported in the literature. Initially, there were great concerns about second impact syndrome and its possible relationship with diffuse cerebral swelling with delayed catastrophic neurologic deterioration. However, the absence of specific risk factors and presence of only scattered case reports (17) make this syndrome and its postulated severe complications controversial and questionable (265). Similarly, the presence of chronic traumatic brain injury, as seen in boxing, has been postulated in football but without significant evidence (349). Occurrence of intracranial hemorrhage, particularly subdural hematoma, has been reported in football (343). Persistent cerebral traumatic injury due to football has been documented in 66 players over the past three decades (65).

Spinal injuries are some of the most devastating injuries associated with football. Next to diving and skiing, football is the most common cause of sports-related spinal cord injury (13% of all sports-related spinal cord injuries) (203). The most common mechanism of spinal injury is axial loading of the cervical spine, as opposed to neck hyperextension or hyperflexion using film analysis of plays with injuries (59). A condition called transient quadriparesis, or cervical transient neurapraxia, seems to be most common in football secondary to a blow to the head with sudden neck flexion or extension. Clinically, the patient experiences transient weakness and numbness to the extremities for a period of seconds to a few days. The incidence of this condition is between 1 and 6 per 10,000 players (444). In children aged 7 to 15 years who developed cervical cord neurapraxia from a football-related injury, the mobility of the pediatric spine, rather than congenital cervical spinal stenosis, may play a role (45), although this is controversial (446). Quadriplegia due to cervical spinal cord injury is predominantly caused by spear tackling (42). Brown-Sequard syndrome has been reported in high school and college football players (42). Over the past decades, a number of changes to the sport, including changes in penalty systems, emphasis on tackling with shoulders as opposed to using the head as a spear, and new designs of helmets and protective equipment, have led to a significant reduction of cervical spine injuries associated with quadriplegia in football since 1976 (446). Obviously, changes are needed, as neck injury rates in football are the highest for sports in the United States, including ice hockey or soccer (98).

Brachial plexus injury is one of the most common forms of injury in football. In Canadian varsity football players, brachial plexus injuries were the third most common specific diagnosis (271), whereas the incidence at two university centers was 49% of all injuries (76). The incidence of plexus injury has been reported to be as high as 2.2 cases per 100 players (77). Initially called a nerve pinch syndrome, this phenomenon is now colloquially termed a stinger or burner. The stinger comprises approximately 36% of all neurologic upper extremity injuries related to football (223). Patients note pain and paresthesias shooting down the arm into a few of the fingers, associated with transient weakness and prompt recovery over minutes. Rarely, weakness may persist for several months (74), suggesting upper trunk brachial plexus axonotmesis as documented by electromyography (76). Electromyography and nerve root stimulation studies may find abnormality in 12% of players with such injury (252). The presence of electromyographic abnormalities in such injuries best correlates with the presence of weakness at 72 hours post-injury (422). The source of the stinger is controversial, with some authors advocating C5 or C6 radiculopathy (341), whereas others believe it to be due to dysfunction of the upper trunk of the brachial plexus (105; 252). More persistent cervical plexus injuries include upper trunk brachial plexopathies and C5 or C6 radiculopathies (222; 223).

Peripheral nerve injuries in football are less commonly reported but may occur due to blocking or tackling techniques. In one study, it was the most common sport to cause injury in patients referred for electrodiagnostic testing (222). Mononeuropathies reported in the upper limbs of football players include axillary neuropathy (333; 222), suprascapular neuropathy, ulnar neuropathy, median neuropathy at the carpal tunnel, long thoracic neuropathy, and radial neuropathy (222). Axillary neuropathy can be associated with shoulder dislocation (222) or can be isolated secondary to direct trauma to the anterolateral deltoid region (333). Although many athletes with axillary neuropathy fail to regain full axillary nerve function, 91% of such athletes return to pre-injury levels of professional sports activities (333). The syndrome of “footballer’s hernia” with lower abdominal bulging may relate to an iliohypogastric neuropathy in some cases (264). Lower limb mononeuropathies in football players consist of peroneal neuropathy, particularly in cases where complete knee dislocation and ligamentous injury has occurred where peroneal neuropathy incidence is 24% (497; 222), and sciatic neuropathy, as a portion of the controversial “hamstring syndrome” (264). An L5-S1 radiculopathy related to football injury was reported in one patient (222).

Golf. Golf is usually regarded as a relaxing recreational sport that would be expected to have a low injury rate. Many of the literature reports regarding golf-related injuries to the nervous system are based on single cases, with a variety of difficulties.

The quick rotation and extension of the neck and head during the golf swing probably has led to vertebral artery dissection, resulting in Wallenberg syndrome (428).

Mononeuropathies in golfers are often atypical. A beginning golfer presented with an unusual location for median neuropathy, with segmental demyelination found 2 cm to 3 cm distal to the wrist crease, after presentation with sensory deficit within the distal median nerve distribution (183). A professional golf instructor was reported with ulnar neuropathy localized as a focal conduction block in the distal forearm approximately 7 cm proximal to the ulnar styloid, perhaps due to enlargement of the flexor carpi ulnaris and subsequent compression of the adjacent ulnar nerve found at surgery (63). Another golfer developed an electrophysiologically proven mononeuropathy of the deep palmar branch of the ulnar nerve related to a forceful grasp of a golf club handle in the left palm (182). A movement disorder called “yips” (not a true injury) has been described in golfers with a jerk, spasm, or other disruption of movement during putting or chipping; this can be considered a focal task-specific dystonia (370).

The use of a golf cart has surprisingly been associated with rare spinal cord injury following serious trauma (207). Head injuries, including cerebral contusions, skull fracture, and extradural hematoma, have each been reported once as a result of a golf cart-related traumatic injury (452).

Head injury due to a golf club (91%) or golf ball has been reported in children aged 3 to 13 years, 78% of whom were boys (246). Four percent of injured boys had evidence of a compound depressed skull fracture, which appears to be the most common injury in the pediatric age group (351). Of pediatric patients with golf-related head injuries, 33% require neurosurgical intervention (351). There are no reports of adults being injured with golf clubs or balls.

Thoracic spinal injuries, including acute thoracic spinal disc prolapse following a golf swing, have been recorded in a male with a decade history of Lhermitte symptom (190), and thoracic spinal osteoporotic fractures in women (114) have been noted.

Gymnastics. Gymnastics requires difficult maneuvers and body postures that place the body at risk for injury, particularly with involvement of trampolines. Specifically, the spine and spinal cord are at substantial risk with gymnastics. Twenty-four former female artistic gymnasts had radiography at the end of their athletic career, still at a young age, which demonstrated radiographic changes in more than 50%, including spondylolysis, spondylolisthesis, retrolisthesis, and scoliosis (217). Although once hypothesized to play a role, the presence of congenital spinal stenosis in children presenting with cervical cord neurapraxia resulting from gymnastics and other sports does not appear to play a role (45). In young adult gymnasts, the most common mechanism of injury may be failure to perform a technically adequate somersault (307).

The number of injuries to the cervical spine among gymnasts in some studies is next to only football and wrestling, with the majority of the spinal cord lesions occurring at the mid-cervical levels (19; 307). Gymnastics is also one of the sports where high school and college participants are at greatest risk of death (285).

A single gymnast was reported with a lateral femoral cutaneous neuropathy after she entered an intensive program of jumping rope, with the nerve injury blamed on repetitive hip flexion and extension (144). Femoral neuropathy secondary to iliacus hematoma or hemorrhage within the nerve sheath may also occur in gymnasts (56; 153).

The trampoline has come under attack in recent years as a cause of injury and spinal cord injuries in particular. In many cases, supervision is present and awareness of the danger of the activity is understood, but injuries continue to accumulate at an increasing and alarming rate. The trampoline is an extremely dangerous piece of equipment that can project people up to 9m in the air, resulting in a long fall either back to the trampoline or even to the ground, where the participant is at risk of landing on their head or neck. The high injury incidence associated with trampolines has been verified in other retrospective studies, also suggesting a dramatic increase in the number of related spinal cord injuries over the past decade (131). Spinal injuries compose 12% of all trampoline-related injuries, with the majority of other injuries consisting of fractures (131). Neck injuries more commonly occur in younger trampoline users (416), but teenagers and young adults have been reported with trampoline-related cervical spinal cord injuries also (408). Many authors have called for a ban on using trampolines due to these high incidences of injury (443; 442; 416; 417; 131; 55). Essentially all injuries occur on privately owned trampolines (99%) (131), and the size of the trampoline (even with a mini-trampoline) does not change the risk of injury (398). Adult supervision is likely to be present at the time of injury (56%), and parents often reported awareness of potential the dangers of trampolines before the injury (73%) (417). Often, the parent reported that the child continued to use a trampoline after injury occurred (55%) (417). Aeroball, a sport played on a trampoline court, has been associated with rare cervical spinal injury as well (411). Outside of spinal cord injuries, a trampoline injury was deemed responsible for a vertebral artery dissection in an 11-year-old boy (475).

Upper-extremity fractures and lacerations have led to ulnar neuropathies in children following trampoline-related trauma (247).

Hang gliding and paragliding. Hang gliding is a sport with obvious risks. As with other sports, the majority of injuries reported occur in 20- to 40-year-olds (127). As would be expected, the majority of all injuries (60%) occur at landing (232). In particular, spinal cord injuries leading to paraplegia or tetraplegia have significantly increased (381), and spinal injuries compose 36% of all injuries (232), with thoracolumbar regions most affected (particularly L1) (75; 135).

Hockey. Ice hockey is a sport of controlled aggression where multiple mechanical forces from the player, competitors, the hockey puck, goal posts, ice surface, unyielding boards, and propensity for fighting can all contribute to injury. Inherent speeds of the puck with a slapshot (190 km/h), as well as the skating speeds (30 to 45 km/h) and the speed of the player who is sliding on the ice (25 km/h) all lead to potential for injury (409). Injury rates in hockey that have been published are high, including a total injury rate of 75 cases per 100 high school-level players, or five injuries per 1000 hours of play (138). Of these injuries, 12% were concussions experienced by 9% of the players (138). Injury rates increase with age in ice hockey and are much higher in males (9:1) under the age of 18 years (180). In American men's collegiate ice hockey, overall injury rates are 4.9 per 1000 athlete exposures (13.8 per 1000 game athlete exposures and 2.2 per 1000 practice athlete exposures), with higher injury rates during games than practices (126). A Danish study of adult hockey players found an equally high injury rate of 90 cases per 100 players per season, or 4.7 injuries per 1000 hours of exposure (199). In these adult players, concussions accounted for 14% of all injuries (199).

Some of the risk factors for injury in ice hockey are distinct from other sports. The more experienced player is significantly more likely to sustain injury. The older, taller, and heavier a player is, the greater the risk of injury as well. Positions most likely to sustain injury were defensemen and wings. As with other sports, the majority of injuries occur during competition. Particular events associated with injury include forechecking, breakout plays (head injury), and backchecking. Illegal activities in hockey were responsible for 26% of injuries (138). There is also a significant inverse correlation between ice size and collision rates in elite hockey, as larger international rinks are associated with fewer collisions and head impacts, although the style of play in international hockey could certainly play a role (479).

Bodychecking remains the most common cause of injury in hockey (115; 116), as child and teenage players in contact leagues are four times more likely to be injured and 12 times more likely to receive a fracture when compared to players in noncontact leagues (200). Rates of injuries sustained by 11-year-old children playing with bodychecking are significantly increased relative to the rates among 11-year-old players without bodychecking (161). Female hockey players have similar overall injury rates (9.19 injuries per 1000 male athlete exposures versus 7.77 injuries per 1000 female athlete exposures) even though intentional body checking is not allowed in female ice hockey (377). Women were more likely than men to be injured via contact with the boards, and women sustained less serious injuries (377). Injuries to children playing ice hockey are predominantly due to checking (57%) (353). In contrast to adult players, only 4% of children’s injuries occur due to illegal activity (353); however, 32% of children state that they would check illegally to win, and 6% said that they would purposely injure their opponent to win (353).

In most studies, injuries within hockey most commonly occur at the head and neck. In a study of male collegiate players, concussions were the most common cause of injury in hockey players (126). Peak accelerations inside the helmet are significantly higher for hockey players than for football players, which likely contributes to the risk of head injury (298). Every season, 10% to 12% of minor league hockey players aged 9 to 17 years old suffer a head injury, usually concussive in nature (30). The rate of concussion rises with the age of the player: youth players aged 5 to 17 years suffer 2.8 concussions per 1000 player hours; University hockey players have a rate of 4.2; and elite amateur players have a rate of 6.6 (177). This strong trend with age is probably associated with greater rates of bodychecking. At the professional ranks, concussion rates are even higher, ranging between 20 and 30 concussions per player hours (480). More concerning is the significant increase in concussion rates over more recent seasons of professional ice hockey. Although this increase may be due to increased recognition and reporting, other changes such as larger and faster players, as well as harder boards and glass may also be a contributing factor (480). Head injuries in ice hockey are most commonly associated with collision with another player (45%), whereas other causes include hitting the boards (34%) or being hit by a stick (22%) (138). Although many injuries in hockey are due to illegal activities, fighting does not appear to be a common cause for concussion, although illegal elbowing may be (145). In some studies, concussion is the most common ice hockey-related injury (177). Even in female hockey players, where intentional body checking is not permitted, concussion was still the most common injury overall (377). Concussions occur in hockey at an estimated rate of 3.7 per athlete season (77), or between 4.6 to 6.0 concussions per 1000 player-game hours (145). The annual risk of concussion related to professional ice hockey is about 5% per player (439). As with spinal injuries in hockey, the frequency of concussion may be increasing, although increased attentiveness and improved diagnostic skills may bias this observation (29). In contrast to other sports again, the incidence of concussion increases with higher levels of play and experience of the player (177). The use of face shields was studied in Canadian University hockey players with comparison of full and half face shields. The use of a full face shield gave a significant decrease in time lost from participation because of concussion and no difference in the rates of neck injury or concussion (27). Other forms of more severe brain injuries, such as epidural and subdural hematomas, are rare but reported (177). There are rare mortality occurrences in an observer or participant secondary to their head being struck by a puck. In one case, a single patient, struck over the left mastoid region by a hockey puck, suffered a fatal rupture of a left vertebral artery berry aneurysm (371).

Benson and associates launched an aggressive study to determine concussion rates among National Hockey League players (26). Over seven regular seasons, this group reported 559 concussions, or 1.8 concussions per 1000 player hours.

Spinal injuries, although less common, are probably the most devastating form of ice hockey-related injury. The great majority of male athletes suffering major spinal injuries are aged 16 to 20 years (433; 437). The most common level of injury has been at the cervical spinal level (85%) (432), whereas another Scandinavian study reported 69% of all vertebral injuries to occur between C5 and C7 levels (279). Spinal injuries occur most commonly with pushing or checking into the back (37% to 40%), whereas impact with the boards accounted for 66% to 77% of spinal injuries (432; 437). The injury suffered at impact with the boards is often a burst fracture of the cervical spine while the neck is slightly flexed (435). Approximately 50% of spinal cord injuries occur in the 16-to-20-year age group, with most occurring in competitive-level games (432).

Peripheral nervous system difficulties are less common in ice hockey participants. Although more common in football players, the stinger has been reported in hockey players (123). Tarsal tunnel syndrome secondary to inflatable ice hockey skates was reported in a male recreational hockey player with significant clinical and electrophysiological improvement after cessation of wearing the skates (472). Peroneal neuropathy has only rarely been documented in hockey players due to either laceration of the nerve with a skate blade or direct blunt nerve trauma (397; 245).

Inline skating, roller hockey, and skateboarding. Although a less common sport, professional roller hockey, or inline hockey, may actually be associated with more injuries than ice hockey, which may be due to differences in playing surfaces or with difficulty in stopping while on inline skates (461). Although a separate study identified similar total injury rates between inline hockey (139 per 1000 athlete exposures) and ice hockey (119 per 1000 athlete exposures), it was noted that games lost due to injuries were greater in ice hockey (8.3 games in ice hockey vs. 6.5 games in inline hockey) (185). Roller hockey may have a smaller incidence of head and neck injuries than ice hockey (185).

Recreational inline skating has become an increasingly popular recreational activity for both children and adults. Injuries related to inline skating tend to occur in boys (61%) with a mean age of 12 years (378; 305). Head and neck injuries comprise 16% of all injuries, with inexperience with the activity appearing to be the most common cause (378). However, head injuries due to inline skating (34%) are significantly less common than with skateboarding (51%) (378; 317), and the severity of injuries is significantly less than skateboarding (317). Due to these concerns, helmet usage has been advocated in all three sports (378). Two children developed bacterial meningitis following basilar skull fractures after in-line skating accidents (390).

Peripheral nerve injury has not been reported due to inline skates, but tight roller skates have been associated with entrapment of the superficial peroneal nerve (104).

Judo, karate, kickboxing, and related sports. The martial arts are systems of fighting, with each form emphasizing unique techniques and strategies. Although these forms of hand-to-hand combat are associated with numerous injuries, neurologic injuries, usually involving the head and neck, are relatively uncommon (455). Approximately 4% of injuries in karate are concussive in nature (18). The incidence of injuries occurring among each of these disciplines is roughly similar, at least between karate, judo, and Muay Thai kickboxing (134). Annual injury rates are estimated to be 2.43 per 1000 for amateurs and 2.79 per 1000 for professionals (134). Adult martial artists are at greater risk of injury than children or teenagers (498). Greater experience (more than 3 years) in the martial arts is also more likely to be associated with injury (498).

Most literature reports of neurologic injury in martial arts participants involve single case reports. One patient developed a "locked-in" syndrome following a cervical trauma that occurred during karate training, speculated to be due to vertebral artery dissection (332). A judo participant developed a cervical myelopathy associated with cervical disc herniation, possibly associated with congenital cervical canal stenosis and judo-related injury (142). Strokes have been reported in martial arts competitors with embolic stroke secondary or possibly secondary to carotid dissection after neck-holding maneuvers or blows to the head or neck (230; 259; 270; 323). Kickboxing and French boxing have also been associated with intracranial arterial dissection (249; 112).

Taekwondo is perhaps the most likely martial art for producing injury (498), as the risk of injury and multiple injuries in tae kwon is three times greater than in karate. Professional taekwondo athletes have an overall rate of injuries of 62.9/1000 athlete exposures (205). In young Taekwondo athletes involved in tournaments, young male athletes had a higher total head and neck injury rate (21.42/1000 athlete exposures) than females (16.91/1000 athlete exposures) (338). Next to contusion, cerebral concussion was the most common form of head injury in both sexes. In professional taekwondo athletes, concussions occur at a rate of 6.9 per 1000 athlete exposures (205). As expected, the most common head injury mechanism was receiving a blow (338). A roundhouse kick is the most common blow to lead to concussion in taekwondo athletes (213). Male adult full-contact Taekwondo competitors suffer slightly higher rates of (7.04/1000 athlete exposures) than child athletes, but the dominant injury mechanism remains the receiving of a blow (6.46/1000 athlete exposures) (337).

More serious injuries within judo have included chronic subdural hematomas (102) and other forms of intracranial hemorrhage (215). The occurrence of dementia pugilistica or other chronic cognitive changes in participants of martial arts has not been recognized (424).

Peripheral nerve injury is unusual in the martial arts, but direct blows leading to a presumed nerve contusion have been reported to affect the ulnar, axillary, spinal accessory, long thoracic, and peroneal nerves (306; 173; 34). A Morton neuroma was reported in one karate participant, presumably due to repeated irritation of the ball of the foot from fighting stances (306).

Motorbiking. Injuries to motorcycle racers are common and often associated with mortality (9% of all injuries) (460). Given that racing motorcycles can reach speeds greater than 306 km/h, this is not surprising (460). Surprisingly, the vulnerability of being on a motorcycle does not lead to injury rates higher than that of automobile racers overall (71). Of all injuries, 10% to 30% are head injuries, and 25% of these head injuries are severe with associated intracranial hemorrhage or mortality (178; 460). Spinal fractures are uncommon, representing 4% of all injuries (178).

Mountain climbing or hiking. Mountaineering is a challenging individual sport where personal goals are often sought. Often, the goal is to ascend the most difficult route with the least amount of necessary equipment. Injury rates among climbers are estimated at two cases per 1000 climbers (388). Obviously, head and spinal injuries are possible, but there is no medical literature-related information on these subjects. Headache, perhaps due to cerebral edema, is a prominent feature of acute mountain sickness (128), as well as dyspnea, weakness, asthenia, and nausea. Headache appears to be secondary to intracranial vascular dilatation due to hypercapnia before the development of hyperventilation due to hypoxia (128). The headache is often clinically similar to migraine. Cerebral edema likely only occurs above 12,000 feet and requires 2 to 3 days to develop (159). The incidence of cerebral edema in all climbers above 12,000 feet may be 1.8%, and it may occur even in experienced climbers (159). Vasogenic edema likely predominates early, with development of cytotoxic edema later (159). High-altitude cerebral edema has documented MRI changes with reversible white matter edema, having a predilection for the splenium of the corpus callosum (160).

Peripheral nervous system injuries are uncommon in mountain climbers and hikers, but reports exist. Tarsal tunnel syndrome caused by repetitive dorsiflexion of the ankle has rarely been reported in mountain climbers (242). The use of a backpack by hikers has been associated with a unique condition called rucksack paralysis, a syndrome leading to injury of the brachial plexus at the upper and middle trunks and occasionally the suprascapular, axillary, and long thoracic nerves (146; 193). Traction on the brachial plexus is the probable etiology, and one predisposing factor for this condition is using a pack without waist support (146). Often, there are paresthesias but no pain. Electrophysiology may demonstrate conduction block or axonal loss in particular patients with rucksack paralysis, with axonal loss suggesting a poorer prognosis for recovery.

Rodeo. Rodeo is another sport with the union of two species: man and beast. Injury incidences vary among the different events but are the highest in bull riding, bareback riding, and saddle bronco events (60). Bull riding, a sport pitting man against bull, has a high injury rate of 32 injuries per 1000 competitor-exposures (60), whereas the overall composite injury rate is 2.3 per 100 competitor-exposures, lower than most contact sports (62). Concussions account for 9% of all reported rodeo injuries (60), second only to knee injuries. Head and neck injuries typically occur during ride or dismount due to violent motions of the animal, whereas concussions typically occur from falling off of the animal or sustaining a blow from the animal (299). In contrast to other sporting events, inexperienced rough stock rodeo competitors have a lower overall rate of injury and severe injury than experienced competitors (61). The only peripheral nervous system injury the author is aware of in rodeo participants was a patient with shoulder dislocation and axillary neuropathy.

Rugby and Australian Rules football. Rugby, or “Rugger,” is an international sport where protective gear is minimal, and aggressive tackling is an integral part of the game. Australian Rules football, or “Footy,” is another aggressive sport that has similarities to rugby and American football and will be considered here. The majority of rugby-related injuries are to the head and neck, and injuries to the cervical spine can be among the most serious injuries. Overall, the head, neck, and orofacial injury rate in Australian football is 2.6 injuries per 1000 participation hours (50). The most common mechanism of injury appears to be cervical spine hyperflexion, often during a scrum, scrimmaging, or tackling, producing fracture dislocations of C4-C5 or C5-C6 (348; 42). The average annual incidence of acute spinal cord injuries in Australian football and rugby players is between 1.5 and 3.2 per 100,000 players. Overall, 39% of injured players became permanently wheelchair-dependent (66). The number of serious spinal injuries had increased over the 1990s but may now be decreasing over the past 1 to 2 decades (376; 42). In one case, a rugby player suffered a fracture and dislocation of the thoracic spine, resulting in paraplegia (466). The phenomenon of transient quadriparesis, most common in American football, has also been observed in rugby (376).

Cerebral concussion is also a concern within both of these sports (314). The concussion rate in Australian football players is 0.49 per 1000 player hours (50). A concerning statistic is that more junior Australian Rules football players have greater concussion rates than more senior players (391). In one study, 23 concussions were recorded over a 20-week football season (263). With video analysis, concussions involved direct head contact most commonly, with the majority of impacts to the temporoparietal region with the striking body segment most typically an arm or shoulder/thorax (268). As found in American football players, neurocognitive difficulties with a digit symbol substitution test correlated with length and number of symptoms post-concussion (263). Although helmet use in rugby players has not been validated, a thick polyethylene foam incorporated into headgear may improve headgear performance (269). Intake of glucose in university rugby players with prior concussion led to enhanced performances on neuropsychological testing, including declarative memory, during tests where pre-glucose testing showed deficit (336). This may suggest that glucose consumption after concussion may influence cerebral neurotransmitter levels affected by concussion and alleviate concussion-induced memory dysfunction.

Australian Rules football players are at risk for obturator neuropathies (48; 49) due to a fascial entrapment of the obturator nerve at the short abductor muscle of the thigh. This condition appears responsive to surgical neurolysis (48; 49).

Although professional Australian footballers have no significant difference in the frequency of headaches when compared to the general population, headaches during the competitive playing season are increased in frequency (266).

Running. Running is a highly repetitive activity uncommonly associated with lower extremity neuropathy. In one detailed assessment of 25 long-distance runners, no signs of neuropathy were found, although mild changes in quantitative sensory thresholds and nerve conduction velocities were reported (111). Peroneal entrapment neuropathies have been reported in serious runners, in one case, bilaterally (253; 277). Peroneal neuropathy in runners has been demonstrated with electrophysiological evidence of entrapment of the peroneal nerve at the fibular neck in several serious runners (233). In one runner, deep peroneal palsy developed due to compression by a mucous cyst at the capitulum peronei identified by sonography and computerized tomography scanning (92). A differential diagnosis of peroneal neuropathy in runners should include the anterior tibial compartment syndrome, an ischemic myopathy presenting with post-exertional pain and swelling and possible foot drop (253). Meralgia paresthetica has also been attributed to jogging (253). Tarsal tunnel syndrome, a compressive lesion of the posterior tibial nerve, can result from repetitive dorsiflexion of the ankle among active runners (242). A branch of the tibial nerve, the anterior calcaneal branch, can become entrapped where the nerve passes between edges of the deep fascia of abductor hallucis and os calcis (171). In runners with heel pain, abnormalities in nerve conduction studies are more commonly found for the medial plantar nerve than lateral plantar nerve (383). Other neuropathies of the foot in runners include those of the interdigital nerves, posterior tibial nerve, superficial peroneal nerve, sural nerve, and saphenous nerve (382). Finally, Morton neuroma can occur in runners and must be differentiated from plantar fasciitis and metatarsal bursitis (101; 468). Even experienced runners are at risk of rhabdomyolysis (396). Excessive load on the heel and midfoot regions may occur in runners with pes planus, which may contribute to neuropathy formation in the feet (418). There are no reports of spinal cord dysfunction in runners.

Skiing, snowboarding, sledding, and ski jumping. Winter sports are some of the most popular recreational activities in the world. Although most winter sport-related injuries are orthopedic in nature, these difficulties can also keep the neurosurgeon busy as well. The incidence of downhill skiing injuries is 2.05 per 1000 skier days (83), and the mortality rate is 1.6 per 1,000,000 skier days. Age groups with the highest skiing-related injury rates are older subjects: the 55 to 64 years (29.0 per 1000 participants), the 65+ years (21.7 per 1000 participants), and the 45 to 54 years (15.5 per 1000 participants) (494). In particular, head and spinal injuries are the most commonly occurring maladies. Of sports-related spinal cord injuries in Germany, 25% of injuries were due to downhill skiing accidents (381). The majority of spinal injuries (70%) caused by downhill skiing result from a simple fall, followed by striking a tree (292; 496). The age group most affected by spinal trauma in winter sports is the 15-to-25-year age group (40%) (136). In contrast to the level of spinal injury in other sports and recreational activities, most skiing-related spinal traumas occur in the thoracolumbar region (47%), followed by the cervical region (39%) (136). The fracture type most commonly associated with skiing-related spinal injury is compression in nature (38%) (345). However, serious injuries to the spinal canal in snowboarding-related injuries occur mostly at the cervical level (136), and 3% of all snowboard-related injuries were spinal in nature (496). Besides the acute presentation of injury, the rare occurrence of a delayed presentation of cervical spinal epidural hematoma 1 month after a snowboarding injury has occurred in a young male snowboarder after striking his occiput and complaining of immediate neck pain and radicular symptoms (216).

Snowboard-related spinal injuries also differ from skiing-related injuries, with a significantly higher incidence of transverse process fractures (496). The initial period of learning how to snowboard appears to be difficult, as beginner snowboarders are significantly more subject to spinal injury than beginner skiers (496). In contrast, intermediate to expert snowboarders are more likely than beginner snowboarders to be injured due to attempting tricks or jumping (496). In one study, young males aged 23 to 25 years suffered spinal cord injury, which in all cases was associated with flexion-distraction type fractures at vertebral junctions, all of which were due to a backward fall from an intentional jump (389). Another study identified intentional jumping as the cause of injury in 77% of snowboarders, as compared to 20% of skiers (431). Greater than half (53%) of all surgically treated spinal injuries of winter sports-related injuries showed some neurologic impairment, whereas 17% had evidence of a complete transverse lesion of the spinal cord (136). The presence of a spinal injury with winter sport activity is often a harbinger of another serious injury, such as craniocerebral trauma (36%) (136). In addition, there was a significantly higher incidence of spinal injury among beginner snowboarders than among beginner skiers. Beginning snowboarders also sustain a significantly higher incidence of emergent injuries, including concussion, as compared to beginner skiers (312). In opposition to older skiing populations with injuries, snowboarding-related injuries are most common in the 10- to 13-year age group (15.9 per 1000 participants), followed by the 14- to 17-year group (15.0 per 1000 participants) and the 18- to 24-year group (13.5 per 1000 participants) (494). In contrast, intermediate or expert snowboarders are more likely to be injured because of jumping than beginners (p < 0.001), whereas about 70% of spinal injuries caused by skiing resulted from a simple fall (496).

Head injuries, although not as publicized as spinal injuries, are common in downhill skiing (294). Traumatic brain injury rates in skiers are highest among older skiers, 55 to 64 years (2.15 per 1000 participants), and very young skiers, 10 to 13 years (1.69 per 1000 participants) (494). Just as with spinal injuries, beginning snowboarders are significantly more likely to suffer a head injury as compared to beginning skiers (294). Head and face injuries compose 17% to 22% of injuries in recreational skiers reporting injuries (248). Approximately 22% of these injuries were related to concussions, and the body region injured most frequently in male skiers (248). In one Canadian study, 60% of skiers presenting after trauma had suffered a head injury (292). Approximately 69% of these injuries are concussion in nature (237). However, 14% of these patients suffered severe brain injuries, with an overall mortality rate of 4% with a severe head injury (237). As opposed to other forms of skiing-related injury, collision with a tree or other object is the mechanism of injury in 47% of skiing-related head injuries. Collisions with trees are more likely to produce severe head injury and mortality. The risk of head injury from skiing is greatest in males and those less than 35 years old (237).

Snowboarders appear to have even higher rates of head injury, up to three times higher, as compared with skiers (172; 237). Competitive snowboarders have an injury incidence rate of 4.0 +/- 0.7 injuries per 1000 athlete exposures (448). Snowboarders are more likely to suffer intracranial hemorrhage with head injury (71%) as compared to skiers (28%) (172). Risk factors for snowboarding-related major head injuries include falling backward (68%), occipital impact (66%), a gentle-moderate ski slope (76%), and inertia (76%) (294; 295). Of cases involving inertia, contrecoup injury was present in 8% of patients (295). Risk factors for acute subdural hematoma as a snowboarding-related injury include falling backward and occipital impact. Subcortical hemorrhagic contusion risk factors include falling during a jump, a temporal impact, or falling on a jump platform (295). In particular, jumping is a much more frequent cause of head injury in snowboarders (30%) than in skiers (2.5%) (129). These snowboarding-related injuries are related to the opposite-edge phenomenon, resulting from a fall on a gentle or moderate slope leading to occipital impact (295). Rarely, skiers and snowboarders can be subject to avalanche, which may be associated with closed head injury (194). Suggestions have been made for snowboarders to wear protective gear over the occipital region and for beginners to avoid jumping (294; 129; 295).

Cross-country skiing, not reported to be subject to head and neck injuries as reported in downhill skiing, has been associated with mononeuropathies. Cross-country skiing has an injury rate of 0.72 injuries per 1000 skier-days, with more injuries occurring in inexperienced skiers, often when they are on a downhill slope (46). An isolated femoral neuropathy was reported in a single cross-country skier with vigorous activity (287), and ulnar neuropathy was reported in another, perhaps due to forceful poling (130).

Although ski jumping would be expected to have high injurious rates given jumps achieving close to 100 meters in height, injury reports are uncommon. Injury rates for non-World Cup and World Cup competitions were estimated at 4.3 and 1.2 injuries per 1000 skier-days, respectively, roughly equivalent to injury rates due to alpine skiing (492). Neurologic injuries are uncommon, with four patients reported to have closed head injuries, including concussions (482).

Injury in short-track speed skating is uncommon, but 6% of elite-level speed skaters are subject to concussions due to accidents during competition. Likewise, junior figure skaters can also be subject to head injuries, which all seem to occur in pairs skating (10% of injuries) and not in single figure skaters (110).

Snowblading is a relatively new sport that uses short, maneuverable skis. Head injury, including concussion, composes 11% of all injuries in snowbladers; it remains unclear if helmet use affects concussion rate or rates of neck injury (54).

Luge is a winter sport where an individual or team rides a sled down a winding ice track. The risk of sustaining an injury in luge is 0.39 per person per year, with most injuries musculoskeletal in nature (87). Concussions compose 2% of all injuries, most of which are due to crashes (87). Recreational sledding has similar degrees of overall injury as compared to skiing but has a significantly higher incidence of head injuries (122), as high as 34% in one study where 3% of injuries were to the spinal column in children aged 18 years or younger (403). Sledding injuries were most common in males aged 5 to 14 years (414). Traumatic brain injuries have been reported in 9% of sledding injuries among children (414). Despite sledding-associated injury rates, helmets are used by only approximately 3% of all childhood sledding participants (316).

Snowmobiling and all-terrain vehicle riding. Snowmobiling is a popular recreational activity, and snowmobiles are used for some competitive racing as well. High speeds (75 km/h) and dangerous riding habits contribute to injuries associated with snowmobile use. Participants injured using snowmobiles are typically males (85% to 90%) and have an average age of 25 to 29 years (120; 24). Helmets are used sparingly (35%), and alcohol intake is present in 44% of cases (24). Serious head injuries compose 34% of all injuries, whereas spinal injuries compose 18% of all injuries (24). Snowmobiling can be associated with a risk of avalanche depending on the setting, and closed head injuries have been reported as a result (194).

Peripheral nervous system injuries due to snowmobiling include brachial plexus injuries in 4.8% of snowmobile accident victims (275). A complete brachial plexopathy is seen in 67% of snowmobile accidents with brachial plexus injury, often associated with orthopedic shoulder injury (53). Supraclavicular injuries were more common and more severe than infraclavicular injuries (275). The author has seen one patient with bilateral ulnar neuropathies at Guyon Canal after a full day of snowmobiling with his hands secured to the handlebars with duct tape (see Clinical vignette for further info).

All-terrain vehicles (ATVs) have become more popular, and injuries due to ATVs have become more common, especially in pediatric populations, which comprise 65% of injured riders (368). Patients as young as 2 to 4 years have been reported to suffer injury from ATV riding (240; 368). Forty-five percent of riders who die due to ATV accidents are under 16 years old (52). Rates of ATV-related injury are highest among males aged 15 to 24 years (4.1 per 100,000), followed by males aged 5 to 14 years (3.8 per 100,000) (52). As with snowmobiling, most injured riders did not wear a helmet (240). The most common ATV-related injuries include skull fracture, closed head injury, intracranial hemorrhage, and spinal fracture (250). The most common mechanisms of injury were falling of the ATV to the ground, striking a tree, or flipping backward (368). Collision with a stationary or moving object is most likely to lead to a fatal injury (52). Next to orthopedic injuries, closed head injuries are the most commonly reported injury (240). In terms of neurologic injuries, cranial injuries compose 64% of the total, and spinal injuries composed 36% (368). Avocation for helmet use has been suggested for ATV users (368).

Soccer. Soccer, a true international sport, has expanded tremendously over the past few decades and is played on the local, regional, national, and international level. Although intentional contact is not extensive, unintentional collisions with other players or goal posts and attempts to strike or head the ball have all led to injuries within soccer. Many of the injuries experienced by soccer players are of a musculoskeletal nature, affecting the lower extremities. However, head and neck injuries can result from collisions,, and rarely from heading the ball. Injury rates in collegiate-level soccer players are estimated at 2.1 per 100 athlete exposures during total events (games plus practices) (350), with 1% of all injuries considered serious. Indoor soccer players have an overall injury rate of 4.5 injuries per 1000 player hours, slightly less than outdoor soccer players (5.6 injuries per 1000 player hours) (115; 116). Female soccer players have a similar overall injury incidence rate of 1.93 injuries per 1000 player hours (141), with much higher injury rates in games (12.63 per 1000 player hours) as compared to practices (1.17 per 1000 player hours) (141).

Given the fast nature of the sport and the ability to “head” the ball, head injuries are the most common form of neurologic injury in soccer. Head injuries have been shown to account for between 4% and 22% of soccer injuries (454; 191). Head injury rates in professional soccer players have an incidence of 1.7 per 1000 player hours, with concussions occurring at a rate of 0.5 per 1000 player hours (12). The action of heading the ball leads to high acceleration forces to the head; in high school-level soccer players, heading of the ball was measured with a triaxial accelerometer and identified significantly higher (160% to 180%) peak accelerations when compared to impacts occurring in football or hockey (298). The most common actions leading to head injury are heading duels (58%); the body parts striking the injured player's head most commonly are the elbow/arm/hand (41%), head (32%), and foot (13%) (12). Indeed, direct heading of the ball seems uncommon to produce head injury (6%) when compared to other forms of contact (339). One study foud that 63% Canadian University soccer players self-reported symptoms of a concussion during the previous year, whereas only 20% of the concussed soccer players realized that they had suffered a concussion (100). Of those soccer players experiencing concussions, multiple concussions were experienced by 82%, suggesting that particular players are more susceptible to head injury (100). Recognized risk factors for concussion include female sex (100). Of all soccer positions, goalies were the players most commonly affected by concussion, even though they are less likely to head the ball (100). Although heading the ball is causative for most reported concussions, goal post injuries have been reported as well (454). More serious head injuries are less commonly reported in soccer. Occurrence of chronic-subacute subdural hematomas with an associated arachnoid cyst and presentation after being struck on the head by the soccer ball has been reported (204; 344). A delayed presentation of epidural hematoma after being struck by the soccer ball has been noted in one report (58). The author has seen one female patient who suffered a concussion and basal skull fracture after being struck by a soccer ball, followed by bacterial meningitis and subsequent mortality.

Although the force of heading the soccer ball and collisions with other players would suggest the possibility of long-term neuropsychological dysfunction, this association is controversial. One study of former soccer players found 81% to have mild-moderate neuropsychological impairment (454). In a study of professional active soccer players, the number of headers occurring in one season was related to poorer results on attentional and visual/verbal memory tests, whereas the occurrence of soccer-related concussions was related to poorer results on attention and visuoperceptual processing tests (255). In another study, a comparison of soccer players with control athletes revealed that amateur soccer players exhibited impaired performance on tests of planning and memory, with the number of concussions incurred inversely related to neuropsychological performance (254). However, neither participation in soccer nor a history of soccer-related concussions was found to be associated with impaired neurocognitive function in high-level soccer players or teenage players in other studies (155; 191). Other studies are skeptical of the previously reported neuropsychological deficits, and it appears doubtful that the subconcussive impact of purposeful heading is responsible for the previously noted deficits. However, the question still remains as to whether or not multiple subconcussive impacts may have some lingering effects (211).

Spinal cord injuries in soccer players appear to be uncommon. The average annual incidence of acute spinal cord injuries in soccer players in Australia is 0.2 per 100,000 players (66). Two male soccer players have suffered cervical hyperextension injuries leading to quadriplegia (407). Peripheral nervous system injury appears to be rare in soccer players. There is a single case report of peroneal nerve compression at the fibular neck attributed to excessive play in only one soccer player (233). A young male amateur soccer player developed a profound facial nerve palsy due to a complex fracture of the left petrous temporal bone after a heading injury (235).

Tennis and other racquet sports. Tennis and other related racquet sports are based on the repetitive action of the swinging arm. This can lead to a number of musculoskeletal difficulties that may mimic a nerve entrapment syndrome. Caution must be used in interpretation of nerve conduction studies in asymptomatic elite tennis players; significant delay in sensory and motor conduction velocities is identified in the dominant arm versus the nondominant arm (79). Specific nerve entrapments occur within tennis players and range from rare to common. Posterior interosseous nerve entrapment is common among tennis players and occurs at the Arcade of Frohse, resulting in weakness of the wrist extensors and metacarpophalangeal extensors (202; 242). Suprascapular neuropathy has also been reported in tennis players, with compression at the suprascapular or supraglenoid notches, as well as with a ganglion cyst (95; 365). Long thoracic neuropathy is rarely reported in the tennis player (468), as well as radial nerve palsy secondary to fibrous arches at the lateral head of the triceps (347).

Volleyball. Of competitive and recreational sports played by high school and collegiate athletes, volleyball is one of the safest (343). Of injuries caused in high school athletes over a 3-year period, volleyball was responsible for only 0.5% of these, again the lowest incidence of all major high school sports (343). Although head and spinal injuries due to volleyball are rarely reported in the literature, peripheral nervous system injuries are reported. One frequent form of mononeuropathy is an isolated entrapment of the suprascapular nerve at the spinoglenoid notch, presenting with painless weakness of dominant arm external rotation with evidence of infraspinatus atrophy on examination (124; 281). This neuropathy only occurs with the serving, or dominant, arm (124). Up to 45% of volleyball players may have subclinical suprascapular neuropathy (124; 113; 176). EMG in these cases discloses denervation and loss of motor units restricted to the infraspinatus muscle with the supraspinatus and other shoulder muscles found to be normal (281). A possible association between increased shoulder joint range of motion and the presence of isolated suprascapular neuropathy has been suggested (487). Alternatively, the medial tendinous margin between the infraspinatus and supraspinatus muscles may impinge against the lateral edge of the scapular spine, leading to compression of the infraspinatus branch of the suprascapular nerve (375). Favoring the latter theory is the favorable response of elite volleyball players to a spinoglenoid notchplasty procedure (375). In those players without response to conservative measures, surgical decompression of the suprascapular nerve may improve symptoms and permit return to play (109). Isolated mononeuropathies of the axillary nerve and long thoracic nerve have also been reported in younger volleyball players, perhaps related to a quadrilateral space syndrome in the case of axillary mononeuropathy (108; 321). Asymptomatic ulnar neuropathy at the elbow is more common among volleyball players than healthy control subjects (319).

Wrestling. Amateur wrestling is an aggressive sport with a frequent number of head and neck injuries. Neurologic levels affected include the brain, spinal cord, and brachial plexus, with many injuries occurring either due to wrestling holds or take down maneuvers. Injury rates in wrestling are second only to football among high school competitors, with an injury rate of 1.58 per 100 player seasons (343). Wrestling also accounts for the second highest rate (10.5%) of mild traumatic brain injuries in collegiate athletes, only second to football (343). Severe injuries consisting of central nervous system trauma within wrestling occur at a rate of 1 per 100,000 participants, with the majority of severe injuries occurring during competitions (80%) (40). An interesting trend of a higher rate of injuries in low- and middle-weight classes was noted (40). Certain positions within wrestling have been more frequently associated with injury, including the defensive position during takedown maneuvers (74%), the down position (23%), and lying position (3%). Severe injuries included cervical spinal fractures (77%), spinal cord contusions with transient quadriparesis (12%), severe closed head injury (8%) and acute lumbar disc herniation (3%), resulting in quadriplegia (33%), residual neurologic deficits (20%), paraplegia (3%), and death due to head injury (3%) (40). Reports of spinal cord injury are not unusual in wrestling competitions (01). The majority of severe head injuries are secondary to head-to-head collision during takedown attempts, but slams to the mat can also result in head injury (419). The rate of concussion in wrestling has been estimated at 2.5 per athlete season in one study (77).

Brachial plexus injuries appear to be relatively common in wrestling compared to other sports and tend to occur with holds when the wrestler’s head goes in a direction opposite to one shoulder, such as with a full or half Nelson hold (419). Burners, or stingers, account for 37% of all head and neck injuries in competitive wrestlers (118; 123). Other forms of peripheral nervous system injury in wrestlers include axillary neuropathy, ulnar neuropathy, carpal tunnel syndrome, long thoracic neuropathy, and suprascapular neuropathy (222). A unique wrestling-related injury occurred with a vertebral artery territory stroke due to a prolonged half-Nelson as reported in a single 17-year-old (364).

The Sumo wrestler may be at particular risk for cervical cord injury due to hyperflexion spinal injuries during takedown maneuvers (293).

The pathophysiology of the particular injury within a given sport is often unique. However, some injuries occur throughout numerous sporting or recreational activities and merit special consideration (Tables 1 and 2).

Table 1. Injuries to the Nervous System, Organized by Sport

Table 2. Injuries to the Nervous System due to Sport, Organized Anatomically (CNS)

Table 3. Injuries to the Nervous System due to Sport, Organized Anatomically (PNS)

Pathological findings in former boxers are intriguing. Neuropathological abnormalities documented in ex-professional boxers include scarring of cerebellar folia with loss of cerebellar Purkinje cells, substantia nigral degeneration, presence of neurofibrillary tangles in limbic grey matter, and cavum septum pellucidum (84; 43). In contrast, CT scanning in amateur boxers has not identified an increased incidence of cavum septum pellucidum (162). Neurofibrillary tangles in dementia pugilistica brains are concentrated in superficial neocortical layers, whereas in Alzheimer disease, they predominate in deep layers (174). Possible markers for neurologic deficit in boxers have been examined in recent years. Professional boxers were scored on the Chronic Brain Injury (CBI) scale, with greater impairment correlating significantly with a greater number of bouts. However, those boxers with an apolipoprotein E e4 allele present also demonstrated a significantly greater tendency towards CNS deficit (197). However, in and of itself, presence of the apolipoprotein e4 allele does not predispose athletes to concussion (221). Studies into the molecular profile of tau pathology in patients with chronic traumatic injury, including boxers, with dementia pugilistica have demonstrated the same tau epitopes as found in filamentous tau inclusions in Alzheimer disease brains. Thus, recurrent brain injury in boxers may activate similar pathological mechanisms as those in Alzheimer disease (379). Measurement of the glial protein S-100B in serum before and after amateur boxing competitions has demonstrated an increase in S-100B protein after boxing. The increase in S-100B protein levels, which has a postulated relationship to cognitive deficits, correlated significantly with the number and severity of the strikes to the head (318). In addition, cerebrospinal fluid sampling in amateur boxers weeks to months after a bout reveals transient elevation of neurofilament light protein and tau, possible markers for neuronal and axonal injury, with highest elevations seen among those receiving the most blows (499) as compared to healthy, non-boxing males.

Recognition of a link between repetitive chronic traumatic brain injury and hypothalamic-pituitary axis dysfunction has been described. Studies of amateur boxers have documented isolated deficiencies in growth hormones (206). This finding has led to further studies showing improvement in lipid profiles and quality of life with physiologic growth hormone replacement in growth hormone-deficient former boxers (430). The same group investigated the presence of antipituitary and antihypothalamus antibodies in active amateur boxers. These antibodies were seen in just over one fifth of the boxers and no normal controls. Pituitary insufficiency was seen in 46% of the antihypothalamic antibody-positive boxers compared to only 10.4% without the antibodies. Antipituitary antibodies lacked the same predictive effect (429).

Epidemiology

The epidemiology of particular injury within particular sports is often unique to that activity. The most common sports have been studied in terms of epidemiology, and some information can be found.

Baseball. Injuries due to more acute causes are probably more common than chronic injuries in baseball. In children aged 7 to 13 years, the acute injury rate per 100 athlete exposures was 1.7 for baseball and 1.0 for softball, with contusions the most frequent type of injury, whereas concussion comprised about 1% of all injuries (350). The frequency of injury per team per season was 3.0 for baseball and 2.0 for softball, with more injuries occurring in games rather than practices (350). At a high-performing amateur level (the 2004 Olympics), 29 injuries per 1000 player matches occurred, with neurologic injuries extremely rare (201). A study of 10 different sports during the 3-year study period identifying mild traumatic brain injuries found that softball accounted for 2.1% and baseball accounted for 1.2% of mild traumatic brain injuries (343). These incidence rates were significantly less than other major childhood sports such as football and hockey. In high school student athletes, the overall injury rate per 100 player seasons was 0.46 for softball and 0.23 for baseball (343). The incidence of mild traumatic brain injury among high school athletes is limited in baseball players, with an injury rate of 0.23 per 100 player seasons, 15 times less than that of football (343). In little league baseball players aged 7 to 18 years, an overall injury rate of .057 injuries per 100 player hours, with a severe injury rate of .008 injuries per 100 player hours, with 46% ball-related injuries and 27% collision-related (324). The most frequent mechanism in little league players was being hit by the ball, representing 62% of acute injuries (324). Catastrophic injury rates in baseball are 0.37 per 100,000 high school player-games, and 1.7 per 100,000 college player-games (41). Fatality rates in baseball players are 0.067 per 100,000 high school athletes and 0.86 per 100,000 college baseball players (41). One of the most common locations for the baseball to strike is the head, and in one study of trauma center hospital assessments of 10- to 19-year-olds, 55% of all baseball injuries involved ball or bat impact often of the head (73).

Basketball. The incidence of mild traumatic brain injuries in high school basketball players is lower than in other organized sports such as football, wrestling, and soccer. The injury rates reported in an observational cohort study per 100 high school player seasons were 1.04 for girls' basketball and 0.75 for boys' basketball (343). In Olympian basketball players, male injuries occur at a rate of 64 per 1000 athletematches, and female injuries occur at a rate of 67 per 1000 athletematches (201), with no indication of neurologic injuries in either group. In one study of trauma center hospitals, basketball injuries capable of causing head injuries in 10- to 19-year-olds were related to striking the basketball pole or rim, or being struck by a falling pole or backboard (73). Reports of professional basketball players sustaining neurologic injuries are conspicuously absent throughout the medical literature, but female professional players have injury rates of 24.9 per 1000 athlete exposures compared to male player rates of 19.3 per 1000 athlete exposures (96).

Boxing. One roadblock limiting the assessment of incidence of more serious injuries in boxing is the potential minimization of injuries due to regulatory policy. It is suspected that the incidence of serious acute head injury in amateur boxing and noncompetitive boxing is lower than in the professional ranks, perhaps due to more regulation and reporting of injuries. In one study of instructional boxing in the United States Marine Corps, only one serious head injury per 60,000 participants occurred, composing only 0.3% of all boxing-related injuries during the study period (366). Another study of amateur boxing found that 0.58% of participants suffered a severe concussion or multiple knockouts (39). In professional boxers, most studies have occurred in New York State and have demonstrated knockout rates of 3% per participant in the 1950s and about three head injuries per 10 boxers in the 1980s (260; 196). An additional study of professional boxing from Nevada has documented an overall incidence rate of injury of 17.1 per 100 boxer matches (36). Male professional boxers are more likely to be injured than female professional boxers, and those boxers who lose by knockout have double the risk of injury of those who do not (36). The incidence of intracranial hemorrhage in boxers is unknown, but acute subdural hematomas are the leading cause of boxing-related mortality (456). Concussion has gained more recognition as a medical condition requiring attention following trauma (208), but attempts to quantify the number of concussions in boxing have not been published. Computerized cognitive assessment in military cadet boxers has demonstrated inferior performances on simple reaction time and continuous performance tests after boxing-induced concussion as compared with baseline testing (469).

Cheerleading. Among children aged 5 to 18 years, the number of injuries per 1000 participants per year ranges from 8.1 for 12- to17-year-olds to 1.2 for 6- to 11-year-olds (399).

Diving and scuba diving. The predominant mechanism of injury associated with water activities is diving, with cervical spinal cord injury composing 4.9% of all water-related accidents in children (186). Among all sports, diving is the most common cause of spinal cord injury (21.6%), and most commonly affects males (88%) of young age (mean of 28.5 years) (203). Patients sustaining a cervical spine injury are more likely to be in the early-mid teenage years (10 years to 14 years in one study) and to be experienced divers (35; 186). The second most common group to suffer high cervical spinal cord injuries due to diving is young adult male divers (154). The cervical spinal cord injury is so common in divers that in one large retrospective study, all spinal cord injuries associated with diving were at the cervical level (381). Almost all spinal cord injuries related to diving (87%) occur in private or residential swimming pools (103). Alcohol and presence of a pool party were respectively involved in 49% and 46% of cases of diving-associated spinal cord injury in a retrospective study (103). As would be expected, most diving-related spinal cord injuries (57%) occur with diving into less than 4 feet of water. Absence of a lifeguard on duty was noted in 94% of cases in a retrospective study. Ordinary diving accounts for 70% of spinal cord injury cases, with unusual or trick dives less commonly associated (103).

Field hockey or lacrosse. In high school lacrosse players, injury rates per 100 player seasons were 0.46 (343). In terms of sports leading to mild traumatic brain injuries, field hockey was the least likely to cause head injury (1.1% of all mild traumatic brain injuries) (343). Lacrosse-related trauma is more common among male players (81% of all cases in a male-dominated sport) and is most common among teenage players (mean age 16.9 years) (106). The overall injury rate for intercollegiate women's lacrosse is 3.8 per 1000 athlete exposures, with a head and face injury rate of 1.4 per 1000 athlete exposures, or 38.5% of all injuries (256). In Olympian field hockey players, injury rates of 55 and 8 per 1000 player matches occur in males and females, respectively, with extremely rare neurologic injuries (201). The one form of neurologic injury that has been studied more extensively in lacrosse players is head injury. The overall head or face injury rate for lacrosse players is 0.71 injuries per 1000 exposures (games and practices) (474). Closed head injuries compose 6% of all lacrosse-related injuries and were slightly more prevalent among females. The rate of concussion is less than 1 per 10,000 exposures (284), and one case of epidural hematoma after being hit by a lacrosse stick has been reported (358). These statistics have led to recommendations for using protective head/face gear (106).

Football. In most studies, football is reported as the sport most likely to be associated with injury, serious injury, as well as neurologic injury. The injury rate for football in collegiate level athletes is estimated at 1.5 per 100 athlete exposures in games as well as practices (350). High school football is no different as it was also associated with the highest injury rate per 100 player seasons (3.66) (343). As would be expected, contact with another player was the most frequent method of injury in football (350). Of all injuries reported, 14% were considered serious (fracture, dislocation, or concussion. The frequency of injuries per professional team per season is 14, more than three times all other team sports (350). Concussions occur in football at an estimated rate of 6.1 per athleteseason in one study (77), more than twice the incidence of other team sports. High school football players self-reported an incidence of concussion of 47% over one season, with 35% of all players reporting multiple concussions (229). Another study of both high school and college football players reported only 5% of players sustained one concussion, whereas 15% of those players sustained a second concussion during the same season (156). The most common specific diagnosis in Canadian varsity players was concussion (271). A study of Canadian Football League professional players suggested a 45% concussion incidence rate over one season, with a 70% incidence of multiple concussions in players reporting at least one (99). In college football players, the incidence of concussion was equally distributed between games and practice (100), which is unique for this sport and injury type as most sports-related injuries have higher incidences during competitions. A slightly increased incidence of concussion was noted among offensive and defensive lineman (100) and special teams players (09). Blocking may lead to more concussions than tackling (09).

The incidence of cervical spine injury in football has fallen over time. From 1971 to 1975, the National Football Head and Neck Injury Registry suggested a rate of 4.14 per 100,000 for cervical fractures and dislocations and 1.58 per 100,000 for quadriplegia (445). In all likelihood, the introduction of modern helmets led to further increases in spinal injuries. Over time, increased protection and rules against headfirst contact led to reductions in the incidence of spinal injuries. From 1976 to 1987, the rate of cervical injuries fell from 70% from 7.72 per 100,000 to 2.31 per 100,000 at the high school level (447). Traumatic quadriplegia also decreased by 82% during the same time period. Data indicate a plateau in the incidence of traumatic quadriplegia. In 2002, the incidence of this traumatic quadriplegia was 0.33 per 100,000 in high school football and 1.33 per 100,000 in college football (296). Catastrophic cervical spine injuries in scholastic football participants per year have an incidence of 1.10 and 4.72 injuries per 100,000 high school and 100,000 college participants, respectively (42). Quadriplegia due to cervical spinal cord injury occurs in 0.50 per 100,000 high school football players and 0.82 per 100,000 college football players (42).

Gymnastics. Trampolines have been held responsible for more than 6500 pediatric cervical spine injuries in 1998 in the United States, a 5-fold increase in reported injuries over the past 10 years (55). This surprisingly high injury incidence has been verified in other retrospective studies, also suggesting a dramatic increase in the number of related spinal cord injuries over the past decade (131). Trampoline-injured patients are both female (53%) and male, with a median age of 7 years (131).

Handball. In Olympian handball players, injury rates of 89 and 145 per 1000 player matches occur in males and females, respectively, with extremely rare neurologic injuries (201).

Hockey. Despite increasing public concern about the number of hockey injuries in both amateur and professional leagues, good epidemiological studies of injury rates and incidence are lacking in ice hockey. In children aged 9 to 16, overall injury rates are 30 injuries per 100 players per season or 4.13 injuries per 1000 player hours (115; 116), with injury rates rising with increasing age. Published injury rates are high, including a total injury rate of 75 cases per 100 high school level players, or 5 injuries per 1000 hours of play (138). Of these injuries, 12% were concussions experienced by 9% of the players. A Danish study of adult hockey players found an equally high injury rate of 90 cases per 100 players per season, or 4.7 injuries per 1000 hours of exposure (199). In these adult players, concussions accounted for 14% of all injuries. In studies where all injuries are considered, the incidence of injuries in ice hockey is astonishingly high, with values between 36 and 66 per 1000 player-game hours (278). Injury rates in ice hockey are much higher in males (9:1) under the age of 18 years (180). In American men's collegiate ice hockey, overall injury rates are 4.9 per 1000 athlete exposures (13.8 per 1000 game athlete exposures and 2.2 per 1000 practice athlete exposures), with higher injury rates during games than practices (126). Every season, 10% to 12% of minor league hockey players aged 9 to 17 years old suffer a head injury, usually concussive in nature (30). The rate of concussion rises with the age of the player: youth players aged 5 to 17 years suffer 2.8 concussions per 1000 player hours; university hockey players have a rate of 4.2; and elite amateur players have a rate of 6.6 (177). This strong trend with age is probably associated with greater rates of bodychecking. At the professional ranks, concussion rates are even higher, ranging between 20 and 30 concussions per player hours (480). More concerning is the significant increase in concussion rates over the more recent seasons of professional ice hockey (480).

Concussions occur in hockey at an estimated rate of 3.7 per athleteseason (77), or between 4.6 to 6.0 concussions per 1000 player-game hours (145). The annual risk of concussion related to professional ice hockey is about 5% per player (439). As with spinal injuries in hockey, the frequency of concussion may be increasing, although increased attentiveness and improved diagnostic skills may bias this observation (29). In contrast to other sports, the incidence of concussion increases with higher levels of play and experience of the player (177). A Canadian registry of spinal injuries sustained while playing hockey has been updated regularly and now includes cases from 1966 to 1996 (432).

A Canadian registry of hockey-related spinal injuries includes cases from 1966 to 1996 (432) of any fracture or dislocation of the spine with or without injury to the spinal cord or nerve roots as well as cases of transient quadriplegia (432). Since 1981, there has been an apparent rise in spinal injuries, although reporting bias, increasing populations of hockey players, and better diagnostic and reporting skills may play a role (432). An average of 17 major spinal injuries occur annually in Canada (433), and there have been six deaths in Canada due to spinal injury over the monitored period (432). Although football has a higher total number of nonfatal catastrophic neck injuries each year, the annual incidence of spinal cord injury with myelopathy is at least three times greater in Canadian hockey than in American football (434; 296).

Wennberg and Tator reported an overall downward trend in the incidence of concussive injuries in the National Hockey League over the past decade with a concurrent increased number of time loss per incident. They postulate that better injury reduction efforts and adherence to stricter return-to-play guidelines account for this discrepancy (481).

Rugby and Australian Rules football. As with ice hockey, concussion rates appear to be rising within Australian Rules football, from 2.2 to 4.7 concussions per 1000 player hours from 1992 to 1996 (313). A concerning statistic is that more junior Australian Rules football players have greater concussion rates than more senior players (391). In one study, 23 concussions were recorded over a 20-week football season (263). Overall, the rate of head, neck, and orofacial injuries in Australian football is 2.6 injuries per 1000 participation hours (50). The average annual incidence of acute spinal cord injuries in Australian football and rugby players is between 1.5 and 3.2 per 100,000 players but may be declining over the past two decades (28). Overall, 39% of injured players became permanently wheelchair-dependent (66).

Skiing, snowboarding, sledding, and ski jumping. Estimated incidences of spinal injury due to downhill skiing and snowboarding are 0.01 per 1000 skier days and 0.04 per 1000 snowboarder days, respectively (431). Of all winter sports-related injuries, spinal trauma composes about 5% (136), with 82% of these due to skiing accidents. Of all skiing-related injuries, 1.4% are spinal in nature (496). The age group most affected by spinal trauma in winter sports is the 15-to-25-year age group (40%) (136). In comparison, 3% of all snowboard-related injuries were spinal in nature (496). Snowboard-related spinal injuries also differ from skiing-related injuries with a significantly higher incidence of transverse process fractures.

The estimated incidence of head injury is 6.5 per 100,000 visits for snowboarders and 3.8 per 100,000 visits for skiers (294). Just as with spinal injuries, beginning snowboarders are significantly more likely to suffer a head injury as compared to beginning skiers (294). Head and face injuries compose 17% to 22% of injuries in recreational skiers reporting injuries (248). Approximately 22% of these injuries were related to concussion, and this is the body region injured most frequently in male skiers (248).

Although ski jumping would be expected to have high injurious rates given jumps achieving close to 100 meters in height, reports of injury are uncommon. One estimation of risk in experienced ski jumpers was 9.4 injuries per 100 participant years (493). Injury rates for non-World Cup and for World Cup competitions were estimated at 4.3 and 1.2 injuries per 1000 skier-days, respectively, roughly equivalent to injury rates due to alpine skiing (492). Neurologic injuries are uncommon, with four patients reported to have closed head injuries, including concussions (482).

The risk of sustaining an injury in luge is 0.39 per person per year, with most injuries musculoskeletal in nature (87). Concussions compose 2% of all injuries, most of which are due to crashes (87). Recreational sledding has similar degrees of overall injury as compared to skiing but has a significantly higher incidence of head injuries (122), as high as 34% in a study where 3% of injuries were to the spinal column in children aged 18 years or less (403).

Volleyball. For high school student volleyball participants, the injury rate is only 0.14 per 100 volleyball player seasons, the lowest of 10 sports examined in one study (343). Of injuries caused in high school athletes over a 3-year period, volleyball was responsible for only 0.5% of these, again the lowest incidence of all major high school sports (343). Although head and spinal injuries due to volleyball are rarely reported in the literature, peripheral nervous system injuries are reported. In one study of international-level volleyball players, the overall prevalence of suprascapular neuropathy was surprisingly found to range from 33% to 45% based on clinical and electrophysiological examination (113; 176). Up to 12% of volleyball players may have subclinical suprascapular neuropathy (124).

Prevention

Prevention of neurologic injuries due to sport should be considered paramount by teams, trainers, physicians, therapists, and the players. Preventative measures from different sports have been associated with possible or definite improvements in injury rates.

Auto racing. One innovation that may improve the safety of race car drivers or racing motorcyclists is the introduction of additional chicanes (turns placed in the fastest part of the racetrack), which decreased the risk of severe injury from 0.1% to 0.03% (236).

Baseball. One intervention in baseball players that has led to relative injury reduction without adverse effect on player performance or player acceptability was the use of a face guard on the batter’s helmet in one nonrandomized study (94).

Bicycling. Recovery in cyclists with ulnar neuropathy may occur spontaneously or with avoidance of activity, rather than with an operative procedure (488). Modification of hand grips on the bicycle handlebars may result in recovery in cyclists with Guyon Canal ulnar nerve compression (166). Cyclists may also develop pudendal neuropathies secondary to racing-bicycle saddles applying pressure on the perineum. Changes in bike saddle position and riding technique may lead to symptomatic improvement (406).

Head injuries due to bicycle riding have become a hot political topic, with many cities in Canada and selected districts within the United States, leading to the institution of mandatory bicycle helmet laws. In one Canadian study, helmet use was associated with less likelihood of hospital admission following injury, a lower incidence of head and facial injury, and a lower incidence of concussion (239). An American study also found that the use of bicycle safety helmets for children led to a lower incidence of skull fractures and may have prevented deaths due to head injuries that occurred in helmetless children (393).

Boxing. Amateur boxing differs from professional boxing with duration of fights, nature of rules and regulatory policies, degree of medical evaluation, and use of protective devices (ie, headgear). Bartsch and colleagues examined the effect of headgear on impacts in boxing and mixed martial arts (MMA) and found significant impact parameters in both sports with significant declines with standard headgear (22). Headgear prevented linear impact dosage superior to rotational damage. Studies like this could perhaps compel licensing boards of professional boxing and MMA to protect their athletes against both immediate and chronic brain injury.

Cricket. Mean match injury incidences are low in cricket, ranging from 48.7 per 10,000 player hours in test matches to 40.6 per 10,000 player hours in 1-day international cricket, with injury prevalences of 11.3% and 8.1%, respectively. In domestic cricket matches, match injury incidence rates are 13.9 per 10,000 player hours for first-class cricket and 25.4 per 10,000 player hours in 1-day domestic competitions (251); batsmen and fast bowlers are most likely to be injured, with neurologic injuries appearing to be very rare.

Diving and scuba diving. Clearly, diving-related spinal cord injury is a sport-related neurologic injury that is best addressed with prevention. An intervention program for subjects with low diving skills suggested improvement and greater safety following participation (37). The type of dive determines risk of injury as well, as dives with a maximized flight distance and low entry angle appear to be safest (38).

Equestrian or horse racing. Closed head injuries are a common occurrence in riders due to falls from the horse. Protective headgear reduces the risk and severity of head injuries. Helmet use should be vigorously promoted (125), as most riders are helmetless (220). Use of a helmet in pediatric riders was associated with decreased frequency and severity of central nervous system injury (44).

Football. Even with the presence of helmets, the high velocity and violent nature of football leads to many head injuries. The presence of a helmet for protection can also lead to injuries, as 7% of all football injuries involve being struck by an opponent's helmet (73).

Although we think of modern helmets as being profoundly more protective than the "leatherheads" used in the early 20th century, Bartsch and associates applied modern biomechanical testing to "leatherheads" and newer helmets and found comparable or even superior profiles of the vintage headgear, suggesting the need for further improvements (23). The use of a technologically advanced Revolution helmet in high school football players was associated with a 31% decrease in relative risk for concussion (80).

Hockey. Bodychecking remains the most common cause of injury in hockey, as child and teenage players in contact leagues are four times more likely to be injured and 12 times more likely to receive a fracture when compared to players in noncontact leagues (200). Female hockey players have similar overall injury rates (9.19 injuries per 1000 male athlete exposures versus 7.77 injuries per 1000 female athlete exposures) despite the fact that intentional body checking is not allowed in female ice hockey (377).

The use of face shields was studied in Canadian University hockey players with a comparison of full and half face shields in the prevention of head and facial injuries. The use of a full face shield gave a significant decrease in time lost from participation due to concussion and no difference in the rates of neck injury or concussion (27).

Spinal injuries, although less common, are probably the most devastating form of ice hockey-related injury. Since 1981 there has been an apparent rise in spinal injuries, although reporting bias, increasing populations of hockey players, and better diagnostic and reporting skills may play a role (432). Rules such as the introduction of "checking from behind" may be more effective at decreasing the number of severe spinal injuries in ice hockey (30). However, programs implemented in Canada in the past two decades have not decreased the number of injuries reported annually (436).

Rugby and Australian Rules football. Despite rugby players' belief that headgear offers protection against concussion (62%), only a minority report using headgear (27%), and only 24% felt that helmet use should be made mandatory (335). Interestingly, a study of childhood and teenage rugby players found that currently used headgear does not provide significant protection against concussion (267).

Skiing and snowboarding. Although head injuries account for 18% of all snowboarding- and skiing-related injuries, using a helmet provides a 60% risk reduction for head injury snowboarders and alpine skiers (427).

Over the past decade, more data have been accrued examining trends in the epidemiology of head injuries with helmets. Russell and associates found a significant reduction in injuries without a concomitant increase in neck injuries (369). Haider and colleagues performed a Cochrane-type review of head injury data and made a level 1 recommendation that all recreational skiers and snowboarders should wear helmets and a level 2 recommendation that the incidence of neck injuries was not increased (164). Cusimano and Kwok performed a similar review and found a 15% to 60% reduction in head injuries and a more significant reduction in head injury involving loss of consciousness (91). However, Sulheim and colleagues, who originally described a 60% head injury reduction, reported in 2016 some attenuation of this effect (426).

Soccer. Injury rates in collegiate-level soccer players are estimated at 2.1 per 100 athlete exposures during total events (games plus practices), with 1% of all injuries considered serious (350). The frequency of injuries per team per season for soccer is three for total events, much less than football (350). Of all injuries over a 3-year period, mild traumatic brain injuries in soccer accounted for 6.2% in both boys’ and girls' soccer at the childhood and high school levels (343). At the high school level, injury rates per 100 player seasons were 1.14 for girls' soccer and 0.92 for boys' soccer (343). In Olympian soccer players, rates of injury are 109 and 105 injuries per 1000 player matches for male and female players, respectively, with neurologic injuries occurring extremely rarely (201). In professional female soccer players, injury rates are considerably higher in defenders (9.4 injuries per 1000 hours) and strikers (8.4 per 1000 hours) than goalkeepers (4.8 per 1000 hours) and midfielders (4.6 per 1000 hours) (121).

Canadian University soccer players self-reported symptoms of a concussion during the previous year in 63% of soccer players, whereas only 20% of the concussed soccer players realized that they had suffered a concussion (100). Of those soccer players experiencing concussion, multiple concussions were experienced by 82%, suggesting that particular players are more susceptible to head injury (100).

Differential diagnosis

Common illnesses may also occur in the athlete. For example, carpal tunnel syndrome commonly occurs in the general population and may also occur in athletes participating in a number of sports. The physician should always consider the coincidental neurologic entity in assessing any athlete with a chronic neurologic complaint.

Diagnostic workup

Whether diagnostic procedures are useful depends on the nature of the injury and the presence of appropriate findings on examination.

In cases of head injury, workup may include, on an individual basis, neuroimaging including CT and MRI scanning, neuropsychological assessment, and in some cases, electroencephalography. Vagnozzi and colleagues have described changes after concussion in athletes on MR spectroscopy (459). They demonstrated temporal lobe changes that persisted after postconcussive symptoms disappeared. Susceptibility-weighted imaging may catch more cases of punctuate hemorrhages compared to normal controls (169). Helmer and colleagues examined both male and female hockey players after concussion with susceptibility-weighted imaging and showed small lesions 2 weeks after concussion (170). This technique is being evaluated as a tool to detect and quantify previously occult lesions (304).

In chronic repetitive head injury, high-resolution MRI is capable of demonstrating a reduction of hippocampal volume associated with impairment in reaction time, but not in cognitive testing (410).

Novel imaging modalities may further elucidate the effects of acute and chronic brain injury in athletes. Wilde and associates evaluated standard MRI findings as well as diffusion tensor imaging of white matter tracts in 10 boxers and compared the results to nine athletes in noncontact sports (484). They also performed cognitive testing. They found defined abnormalities in diffusion tensor imaging in the boxers with neurocognitive deficits, indicating white matter track injury deficiencies in memory and response speed. Hart and colleagues performed MRI imaging of 34 retired NFL football players compared to controls (168). Fourteen of the football players had cognitive issues that correlated with white matter abnormalities and changes in regional cerebral perfusion. Multani and colleagues performed a similar study in 18 retired football players and found a significant association between damage in the superior longitudinal fasciculus and impaired visual learning ability (288). Changes in the corticospinal tracts and anterior thalamic radiations were also seen in the athletes. Coughlin and associates compared PET scanning with a marker of glial cell response in active or recently retired professional football players compared to controls (85). The football players showed a higher total distribution of glial cell activation compared to controls. Because these individuals did not yet show cognitive impairment, it was indeterminate whether these changes would be related to delayed neurocognitive abnormalities.

In athletes with suspected spinal injury, workup may include neuroimaging such as CT and MRI scanning as well as somatosensory evoked potentials in rare circumstances. Assessment of sphincteric function may be necessary in some circumstances. Abbreviated neuropsychological batteries, such as the King-Devick test and the Military Acute Concussion Evaluation, the ImPACT test, and others like them, have been used for rapid assessment by trainers for on-the-field or in-the-ring injuries (06; 132). Galetta and associates described successful use of the King-Devick test to quickly assess impairment in boxers and mixed martial arts fighters (132).

Moving forward, using biomarkers to evaluate injury may become commonplace. Graham and colleagues showed a rise in serum S-100B, neuron-specific enolase, and cortisol in boxers sustaining head blows as opposed to those receiving only body blows (150). Neselius and associates examined both serum and CSF levels of neurofilament heavy (pNFH) and amyloid precursor proteins after amateur boxing matches and found increased pNFH level elevation after subconcussive head injury (302). Levels were not influenced by apolipoprotein subtypes. La Fountaine and colleagues have shown changes in serum prolactin levels from injury through recovery (228). Shahim and associates examined total tau, S-100 calcium-binding-protein B, and neuron-specific enolase plasma and serum levels after concussive injury in ice hockey players and found statistically significant changes in total tau and S-100 but not neuron-specific enolase (395). The same group used serum tau-C levels to demonstrate concussion and serum tau-A levels to demonstrate recovery from concussion and predict return to play after ice hockey injuries (394). Meier and colleagues examined blood markers and concussion prospectively and found significant elevation in S100 calcium-binding protein beta and ubiquitin C-terminal hydrolase early in the course of concussed American football players compared to preseason measurements (272). Oliver and associates performed serial blood monitoring for tau proteins and neurofilament light polypeptide over the course of a college football season and saw distinct difference between starting players whose levels trended up over the course of the year and nonstarters whose levels diminished (311).

Emerging evidence has implicated the presence of apolipoprotein E2 or E4 polymorphism in patients with chronic traumatic encephalopathy. If this link is eventually proven, it could help stratify risk for protective participants in contact sports. However, evidence remains contradictory. Shadli and associates found no influence of apolipoprotein genotype on recovery from mild to moderate head injury (392). However, Tierney and colleagues found a significant association between presence of these alleles and the history of a concussion and reporting of multiple concussions, specifically in college athletes (441). Alosco and associates examined spinal fluid proteins in retired professional American football players compared to age-matched controls (07). They found increased levels of both markers of microglial activation (sTREM2) and elevation of proteins associated with neuronal breakdown (t-tau) in the athletes.

In athletes with suspected plexus, radicular, or neuropathic injury, the use of nerve conduction studies are often complementary, as well as electromyography. Neuroimaging of larger nerves, such as the sciatic nerve and the plexi, can be useful in particular circumstances, especially in cases of acute injury where a hematoma may be suspected.

Diving and scuba diving. Cerebral infarction has been reported in professional breath-hold divers during repetitive dives, with MRI demonstrating multiple T2-weighted hyperintensities corresponding to their neurologic deficit (214). The most sensitive, but nonspecific, diagnostic testing procedures for patients with neurologic decompression illness are EEG and MRI, although there is controversy regarding the presence of significant differences from controls (291; 413). In divers with arterial gas emboli, MRI of the brain identified ischemic cerebrovascular lesions in 75% of patients, as compared to focal hyperintensities in 25% of divers with decompression illness only (354). MRI of the spinal cord is less sensitive in divers with decompression illness, with focal hyperintensities identified in only 14% of patients (354). Somatosensory evoked potentials have identified latency abnormalities in 81% of patients with decompression illness (290). Brainstem auditory evoked potentials have demonstrated prolonged cochleopontine and cochleomesencephalic transmission times after diving in patients with clinical symptoms and signs of vertigo, postural and intentional hand tremor, ataxia, and opsoclonus. These features have also been referred to as otic barotraumas (243; 303).

Football. A future tool in the assessment of concussion-related cognition deficits is fMRI using blood oxygen level-dependent signals. Symptomatic concussed athletes may demonstrate task-related activations in some, but not all, of the expected brain regions. They also may show additional increases in blood oxygen level-dependent activity outside expected regions (72).

Soccer. Electrophysiological evidence of abnormalities due to heading a soccer ball is improbable; nonspecific abnormalities on EEG were found in about 33% of former and active soccer players, although EEG changes seemed to be more prevalent among players who considered themselves “nonheaders” as compared to “headers” (454). A noncontrolled study suggested that of former soccer players, 33% were described as having mild central cerebral atrophy identified with CT scan of the head (454).

Management

The management of individual types of nerve injuries, entrapment neuropathies, and cranial injuries is not specific for or particular to the activity. Rehabilitation for high-level athletes may be particularly intensive. Guidelines for return to competition after concussion have been developed by several different authors--incorporating the magnitude of the initial impairment, number of events within a given time interval, and persistence of symptoms.

McCrea and colleagues examined a cohort of 635 concussed high school and college athletes over a 6-year period from 1999 until 2004. They found that there was a susceptibility to repeat concussion within 10 days of the initial injury, with 79% of repeat concussions occurring during this interval. Interestingly, having a symptom-free waiting period did not reduce the incidence of repeated concussion (261).

Multiple attempts have been made to quantify the severity of concussion and, subsequently, determine the duration the athlete must restrain from competition. Concussions are graded as mild, moderate, or severe based on the duration of loss of consciousness, presence or absence of posttraumatic amnesia, and impairment in the Glasgow Coma Scale Score (64; 342). Return to competition scales factor in more time loss for more severe injuries (return when asymptomatic for mild concussion, 1 week for moderate, and 1 month for severe) and repetitive injuries (64; 342). Unfortunately, multiple concussions have ended or curtailed the careers of many prominent hockey and football players (Steve Young, Troy Aikman, and Pat LaFontaine to name just a few); however, the notoriety of these athletes has also advanced the case of concussion recognition and treatment.

Newer evidence suggests that the development of overt signs and symptoms of concussion may not be necessary to engender pathological changes in cerebral tissue. Gysland and associates used head impact data to quantify the number of subconcussive insults in a college football season and found as many as 1000 such hits in a season (158). Bailes and colleagues reviewed this topic and described athletes with posttraumatic encephalopathy without a known history of concussion, fueling speculation about the deleterious effects of repetitive minor insults on brain function (20). Pratile and associates reported on a cohort of 1213 athletes ages 8 to 18 years who suffered sport-related concussions. When a diagnosis of concussion occurred within 10 days, there was a mean duration of concussion management of 23.5 days. Conversely, when the diagnosis was delayed, mean concussion management was statistically significantly delayed to 37.1 days, suggesting a benefit of early detection and treatment (346).

Wiebe presented the findings of a large Ivy League and Big Ten Consortium studying concussion in athletes. This was a prospective cohort study identifying concussion within 24 to 48 hours and initiating physical and cognitive rest. A total of 1715 concussions were noted, with about two thirds in male athletes and one third in female athletes. About two thirds of the athletes (67.9%) were placed on initial physical and cognitive rest. Amongst these individuals, median return to full academics was 6 days; there was a median 8 days to symptom resolution and median 9 days to return to exertional activities. Athletes who returned to academics while still symptomatic had a higher likelihood of prolonged symptoms and prolonged return to exertion. Predictably, return to exertion before full symptom resolution led to prolonged symptomatology as well, highlighting the need for physical and cognitive rest until symptom resolution (476).

Moving forward, biomarkers may also help assist in returning to play after a concussion. Oris and associates looked at seven biomarkers in the blood of rugby players after a concussion. Biomarkers were collected before and during the season and then after the concussion. Specifically, S100 calcium-binding protein B was most predictive of the resolution of symptoms after concussion. Further research may help guide return to competition as subjective symptomatology is always prone to bias (315).

Long-term consequences of concussion or repeated concussion merit consideration of mental health issues. In a review of the influence of concussion on mental health issues, Gouttebarge and Kerkhoffs relate a 26% risk of anxiety, depression, or both after concussion (149). In a questionnaire returned by 576 former professional athletes, individuals suffering three or more concussions demonstrated a three times greater incidence of depression, and elite athletes suffering at least six concussions had a five times increased risk of depression (148; 149). Rice and colleagues provided a Cochrane analysis looking at 27 studies of mental health issues in former athletes (356). The majority of the studies focused on depression with both short and long-term depression seen after concussions without any clear-cut statistically significant conclusions. Preinjury symptomatology was related to postconcussion persistence of depression and anxiety. Many of the studies were limited because symptoms were self-reported rather than documented by a blinded observer, necessitating further study.

Efforts have been made to replicate return to competition criteria as well as management criteria for spinal injuries and athletes. Schroeder and colleagues address this issue in a consensus paper from the Cervical Spine Research Society (384). Team physicians from the National Football League participated in the research as well. The group reached several conclusions, including recommending a screening MRI study before participation in athletes with a history of cervical spinal injury; ability to return to play in asymptomatic individuals without cord signal change and with normal spinal canal diameter or those with resolved cord signal change and greater than 13 mm core diameter; and ability to return to play in asymptomatic individuals without cord signal change after 1 or 2 level anterior cervical fusion but not 3 level anterior cervical fusion for pseudoarthrosis. The same consensus group has provided valuable reviews on the management of acute spinal cord injury in the setting of collision sports, including on-field management and appropriate transfer and workup (325).

The same organization has also written a consensus statement on lumbar stress reactions occurring from collision sports in the pedicle and pars interarticularis occurring from microtrauma (458). Pain is generally reproduced with extension and can be identified on MRI. Nonoperative care was advised for up to 12 months, with consideration of surgical management if symptoms persist. Recommendations for return to play in asymptomatic athletes after rehabilitation are provided as well. A greater understanding of the long-term effects of repetitive traumas to the cervical and lumbar spine and whether there is a spinal equivalent of chronic traumatic encephalopathy remains elusive.

There are a growing number of cases of thoracic outlet syndrome (TOS) recognized in athletes, with baseball pitchers and cricket bowlers specifically at risk. When thoracic outlet syndrome is documented with radiological and clinical confirmation, both nonoperative and surgical techniques are available. Warwick and Davis described non-operative care of thoracic outlet syndrome in athletes, including physical therapy and both steroids and hydrodissection to alleviate scar tissue (470). Shutze and associates examined surgical outcomes in competitive athletes who have undergone surgery for thoracic outlet syndrome while active in their careers (405). Of 67 patients who responded to a long-term survey, 66% participated in softball or baseball, with volleyball the next most likely athletic pursuit to cause symptoms. Eighty-two percent had resolution of symptoms, and 94% were able to return to activities of daily living without limitation. Seventy percent returned to similar or better levels of athletic activity, with younger age being a better prognostic indicator. Arnold and colleagues looked specifically at major-league baseball pitchers and found 26 individuals who had surgery, with 81% eventually returning to play, with no difference in postoperative career length compared to controls and no difference in postoperative performance (17). Gutman and associates performed a similar review of 27 major-league pitchers undergoing thoracic outlet syndrome surgery and analyzed pitching metrics (157). Twenty of these individuals had neurogenic thoracic outlet, and seven had venous thoracic outlet syndrome. Twenty of the pitchers returned to play in the major leagues with a mean recovery time of 297 days. Earned run average worsened after treatment, although fastball velocity did not.