Long-Term Neurological Consequences Related to Boxing and American Football: A Review of the Literature

BOXERS

In 1928, Martland [1] reported on a condition known as “punch drunk” that was well known to occur in older boxers. A fight promoter acquaintance gave Martland the names of 23 fighters that he considered to be “punch drunk”. Dr. Martland examined 5 of them in person and reported in detail his findings in one case. This 38-year-old ex-boxer had a staggering, propulsive gait and a mask-like facies similar to that seen in parkinsonism, stammering, hesitant speech, and a fine tremor of both hands. His intelligence was normal. Martland gave abbreviated information on the other 4 cases he had personally examined: one had parkinsonian signs, one dragged his legs, another dragged his legs and had slow speech, and the fourth dragged his legs, had slow speech, and slow thinking. Based upon these 5 cases and the descriptions of other “punch drunk” cases by fight promoters, trainers, and fans, Martland stated that the early clinical features began in the extremities with “flopping” of one or both lower extremities and mild unsteadiness occasionally accompanied by “slight mental confusion”. These signs might then progress to distinct dragging of the lower extremities, slowing down of all movements, hesitant speech, and tremors. In severe cases, parkinsonian signs, ataxia, and tremors resulted in a “staggering, propulsive gait”. In end stages, marked mental deterioration may result in commitment to an asylum. Although Dr. Martland was a pathologist, no brain pathology of any boxers was presented. Martland hypothesized that this condition was the result of the accumulation of numerous traumatic brain microhemorrhages sustained while boxing. He presented the pathology results of a case of multiple brain hemorrhages resulting from head trauma in a non-boxer from a large series of such cases he had seen in his medical examiner practice.

Over the next 40 years, a number of clinical case and small series reports of a chronic neurological syndrome seen in retired boxers appeared in the literature. These are well summarized in a table in AH Roberts’ 1969 monograph [2]. These consistently described a clinical neurological pattern of ataxia, slurred speech, cerebellar, extrapyramidal (parkinsonian), and pyramidal signs. There was no mention of suicidality and only occasional mention of depression. Personality changes such as euphoria, aggressive behavior, and sometimes violence occurred on occasion. Memory impairments and dementia also were reported in some cases. It is important to note that when depression, other personality changes, memory problems, or dementia occurred, they never did so in isolation; in other words, these only were seen in cases with coexisting abnormal patterns of clinical neurological signs. To illustrate these points, some of the well-known reports of the neurological abnormalities in retired boxers in the literature between 1928 and 1968 are described in detail.

In 1937, Winterstein [3] reported on the clinical findings of 50 professional boxers that he had examined. He did not detail the individual results but gave his impressions of the group as a whole. He noted a chronic neurological pattern which began with disequilibrium progressing to an unsteady ataxic gait, a “positive” Romberg sign, slurred, hesitant speech and a “vacant look” He also reported that “most of those badly affected” showed bad memory, impaired intelligence, and “mental dullness.” Some of them were “euphoric, some paranoid and suspicious”. There was no mention of depression or suicidality.

In 1957, Critchley [4] reported on 69 cases of chronic neurological disease in boxers he had personally examined. He described the clinical features of what he termed “chronic progressive traumatic encephalopathy of boxers”. He described a gradual evolution of a “fatuous or euphoric dementia with emotional lability” with little insight, progressively slower speech and thought processes, and memory deterioration. The mood was commonly cheerful but “sometimes there is depression”. Irritability and sometimes violent behaviors were noted in some cases. All of these cognitive and/or behavioral issues occurred in boxers with accompanying neurological abnormalities—“almost any combination of pyramidal, extrapyramidal and cerebellar signs”. Dysarthria and tremors were commonly present. Critchley reported that within this “punch-drunk state” there were 4 commonly recurring syndromes: 1) signs reminiscent of neurosyphilis, 2) signs resembling multiple sclerosis, 3) signs resembling those of a frontal brain tumor, and 4) pallidal and striatal signs resembling parkinsonism. Critchley’s impressions were that this syndrome occurred more frequently in professional fighters than in amateurs, more often in 2nd or 3rd rate fighters than in champions, more often in sluggers than in scientific skilled pugilists and more often in smaller compared to bigger men. He believed that the total number of fights and the number of times the fighter had been knocked out were also related to the development of this disease.

In 1959, Neubaerger et al. [5] reported on 2 cases of their own with clinical and neuropathological findings in former boxers and referenced 2 other cases [6, 7] with clinical and neuropathological findings in former boxers. Clinically the cases had neurological signs of extrapyramidal, pyramidal, and cerebellar dysfunction with dementia. Pathologically, three of the brains had increased large astrocytes, neuronal loss, widespread “fibrillary” changes, and cerebral atrophy, and the fourth brain [6] showed extensive plaque formation and cerebral amyloid angiopathy which may have represented early onset Alzheimer’s disease (AD) unrelated to boxing.

In 1962, Spillane [8] reported neurological and pneumoencephalogram (PEG) findings in 5 former professional boxers. He found:

Case 1 had a progressive dysarthria, urinary incontinence, spastic-ataxic gait, hyperreflexia, and a left Babinski sign without personality change or intellectual impairment; PEG showed an absence of the septum pellucidum.

In 1963, Mawdsley and Ferguson [9] reported the clinical features of 8 former professional boxers. They found a pattern of dysarthria, nystagmus, limitation of upgaze, ataxia, tremors, and signs of pyramidal and extrapyramidal dysfunction. All of these cases plus 9 others were reported on by Johnson in 1969 [10], who reported on the clinical neurological and psychiatric findings of these 17 boxers. One case of an amateur boxer who was seen for transient self-resolving anxiety due to family issues was excluded from the results. The clinical neurological “manifestations” of these boxing related syndromes appeared before any psychiatric signs. These included varying combinations, of ataxia, intention tremors, dysarthria, pyramidal and extrapyramidal signs. Of the 16 cases, 11 had pyramidal signs, 7 had extrapyramidal signs, and 11 had cerebellar signs. Johnson found the following psychiatric syndromes in these 16:1) an amnestic syndrome in 11 and dementia in 3, 2) a morbid jealousy syndrome in 5, 3) psychosis in 5, and 4) a personality disorder with rage reactions in 4. In 3 of the cases with rage reactions, impulsive aggressive behavior was a lifelong trait. Johnson did not report any cases of depression or suicidality.

In 1968, Payne [11] reported on the clinical neurological findings and neuropathology of 6 former professional boxers. One case had a manic-depressive psychosis. Another had dysarthria, ataxia, a left 3rd nerve palsy, and papilledema. He had a history of compulsive gambling, sometimes became paranoid and violent, and at one point had set his house on fire. Another had a history of headaches, poor concentration, urinary incontinence, and depression. Examination revealed dementia, dysarthria, and incoordination. Another had dysarthria, ataxia, nystagmus, and incoordination along with headaches, depression, and difficulty concentrating. The 6th case had a normal neurological examination. Neuropathological findings were ventricular dilatation in all 6, cavum septum pellucidum in all 6, fenestrations of the septum pellucidum in 3, and multiple areas of micro scars in the gray matter in all 6.

The most detailed and comprehensive report to date on the clinical neurological features and the epidemiology of chronic traumatic encephalopathy of ex-professional boxers is AH Roberts’ landmark 1969 monograph [2]. Before detailing his findings, Roberts reviewed more than 53 reported cases of chronic neurologic findings in former boxers in the medical literature since Martland’s 1928 paper [1]. He noted that these revealed a pattern of various combinations of dysarthria, ataxia, tremors, parkinsonian signs, pyramidal signs, and cerebellar signs. The bulk of the monograph consists of neurological clinical reports on 224 randomly selected British former boxers. Working with the British Boxing Control Board, the records of 250 (representing about 1.5% of the total number of boxers registered with the board in past years) retired British boxers who had been licensed by the Board for at least 3 years in the past were selected for the study. These 250 and their wives were traced and invited by letter to participate in the study. When necessary, social workers or former boxing colleagues contacted the subjects to encourage their participation. Ultimately, 180 subjects came in to be examined, 43 were examined in their homes, and 1 was examined in the mental hospital where he was residing (total of 224). The examinations performed by Dr. Roberts (a neurologist) consisted of a detailed medical history, boxing history, family history, social history and educational history, and a comprehensive medical and neurological examination. Many, but not all, of the subjects were given “abbreviated versions” of psychometric tests of intellectual function. 37 of the 224 (17%) of the ex-boxers had “evidence of lesions of their central nervous systems similar to those reported and attributed to boxing by others”. Not included in that 37 were subjects who had only isolated neurological signs (e.g., Babinski sign in isolation) and those whose neurological signs were indicative of clinical neurological entities seen in “routine neurology practice.” Roberts presented in detail the findings in 11 characteristic cases. The clinical syndrome characteristically consisted of varying combinations of dysarthria, cerebellar, pyramidal, and extrapyramidal abnormalities with one or more of these predominating in different subjects. Dementia frequently accompanied these findings. Depression was only mentioned in one of the 11 cases described in detail. According to Roberts, this composite picture was “unlikely” to represent a “fortuitous” occurrence of neurological disease in boxers. Further evidence that these neurological abnormalities are related to boxing consists of evidence that their presence was statistically related to the subject’s length of boxing career and number of professional fights in one’s career (but interestingly, not to the number of times the boxer had been knocked out during his career). The limited neuropsychological test results revealed difficulties with synonym selection and poor verbal facility, but memory test results varied with age, not length of career. Sixteen of the 37 subjects with findings of traumatic encephalopathy showed “some degree of impaired intellectual function or personality change”: 2 were definitely demented, 9 had severe memory impairments, 5 had “slow mentation”, and 6 were irritable apathetic or uninhibited. When considering the entire cohort of 224 ex-boxers, 1 had a longstanding history of depression, 3 had minor phobias, and 5 had paranoid states (2 who were demented included in the 37 affected subjects, 2 with psychotic paranoid delusions included among the 37 affected subjects, and 1 with catatonic schizophrenia who was examined in a mental hospital and had no abnormal neurological signs). There was only 1 case among the 224 with clear evidence of a severe memory deficit without any clinical neurological abnormalities on examination.

With the clinical picture of chronic neurological disease of boxers having been well defined, Corsellis et al. [12] completed the picture with a seminal description of the associated pathology in 1973. They reported on the neuropathology of the brains of 15 ex-boxers (12 professional, 3 amateurs) which they had collected over a 16-year period. Their medical and boxing histories and results of clinical examinations were obtained posthumously from interviews with families and review of hospital/medical records. One subject had attempted suicide at age 65 (not completed) and died at age 72. He had parkinsonian features and parkinsonism documented in his medical records. A few of the other subjects had histories of aggressive behaviors and/or paranoia in conjunction with various combinations of dementia, dysarthria, cerebellar, pyramidal, and extrapyramidal signs on clinical examinations. Corsellis et al. described a tetrad of pathological findings that characterized the brains of these ex-boxers:

abnormalities of the septum pellucidum (two subjects had died of intraventricular hemorrhage; in both cases the leaves of the septum were torn and separated but the authors were unsure if this had been accentuated by the bleeding so they did not include those cases in discussion of septal abnormalities). 12 of 13 had a cavum septum more than 3x the width of the small cavum septums they had seen in the brains of non-boxers; the one case without a cavum septum had fenestrations of the cavum; 11 of the 13 had gross fenestrations of the cavum;

The authors stated that this tetrad of abnormal neuropathological findings cannot be explained by any condition/disease other than exposure to boxing. They also pointed out that there was clinic-pathological correlation between the neuropathology they were reporting and the clinical features of chronic brain disease in ex-boxers. Clinical disorders of memory/dementia correlate with neuropathology of the limbic areas (septal regions, fornix, mamillary bodies, medial temporal gray regions), rage reactions and other “abnormal affective states” correlate to these same regions, parkinsonian signs correlate to neuropathology in the substantia nigra and cerebellar signs correlate to the scarring/Purkinje cell loss in the cerebellar hemispheres. There was now a well-defined clinical picture associated with a highly correlated well-defined neuropathology of the disease that can be called chronic traumatic encephalopathy of boxers.

In the 1980 s and 1990 s, there were a few studies employing new neuroradiological techniques (i.e., CT scanning), more standardized neuropsychological testing, and newer brain microscopic staining methods to further refine knowledge of chronic brain injury in boxers. In 1982, Kaste et al. [13] reported the results of clinical neurological, neuropsychological, and brain CT scan testing on 14 Finnish retired champion boxers (8 amateurs, 6 professional) ages 19–53. One professional had apraxia, unsteadiness, and slowed mental functions on neurological examination. He and another professional reported episodes of “embarrassing inappropriate behavior”. On neuropsychological testing, 12 subjects had abnormal results on Trail making tests and 2 of the professionals had abnormal results on more than 1 of the other subtests that made up the battery. On CT scans, 3 of 6 professionals and 1 of 8 amateurs had cerebral atrophy; 2 professionals and 1 amateur had a cavum septum pellucidum (CSP).

Also, in 1982, Casson et al. [14] reported on CT scans of 10 active professional boxers; 5 of the 10 had cerebral atrophy and 1 had a CSP. In 1983, Ross et al. [15] reported the results of CT scans and neurological examinations of 40 ex-boxers (amateurs and professionals) ages 21–73. Of the 38 who had CT scans, 20 had cerebral atrophy (presence or absence of CSP was not evaluated). The presence of cerebral atrophy was directly correlated with the number of total boxing matches fought by the subjects. Six of the 24 boxers who underwent clinical neurological examinations had abnormalities including memory loss, ataxia, difficulty with tandem gait, diminished deep tendon reflexes, and sensory abnormalities.

In 1984, Casson et al. [16] reported the results of neurological examinations, CT scans, and neuropsychological tests in 18 modern (all fought in the post-World War II era) boxers (13 retired professionals, 2 active professionals, 3 active amateurs) ages 18–60. The retired professionals were specifically selected from a group of 29 who were still associated with the sport in New York. Twenty-three who were age 60 or younger, had not retired from boxing for medical, neurological, or psychiatric reasons, and had no known history of neurological or psychiatric disease and no known history of alcohol or drug abuse were contacted and asked to participate in the study. Thirteen former boxers volunteered and were included in this report. All were fully employed. Such a selected group “would not be expected to have obvious symptoms or signs of neurological dysfunction”. Eight of the 15 professionals had cerebral atrophy on CT scan and 3 of these 8 also had a CSP. All 3 amateurs had normal CT scans. Five of the 13 retired boxers had abnormalities on neurological examination (3 dementia including 1 who also had a unilateral Babinski sign, 1 impaired recent memory without dementia, 1 dysarthria and nystagmus). All the active professionals and amateurs had normal neurological examinations. Every subject had more than one abnormal neuropsychological subtest score (see Chart 1 for details). The group as a whole performed especially poorly on tests of recent memory, timed tests, and tests of executive function. Abnormalities on CT scan and neuropsychological testing correlated with number of professional fights but not with the number of knockouts or episodes of amnesia sustained.

In 1990, GW Roberts et al. [17] reinvestigated the neuropathological slides from Corsellis et al.’s 15 cases [12] with recently developed “immunocytochemical methods and an antibody to the beta protein present in Alzheimer Disease plaques”. They reported that all of the cases “showed evidence of extensive beta protein immunoreactive deposits” that had not been detected by the staining techniques that were available to Corsellis et al. in 1973. The authors recommended that Corsellis et al’s finding of tangles but no plaques be amended to acknowledge the presence of “substantial beta protein deposition”. This paper was a purely neuropathological investigation with no clinical data. In the first paragraph of the paper, however, the authors mentioned 3 clinical stages of the “punch drunk” syndrome: 1) affective and psychotic symptoms; 2) social instability, worsening psychiatric symptoms, memory loss and parkinsonism; and 3) the final stage of worsening cognition progressing to dementia and pyramidal tract disease.

There is no footnote reference for this statement and no data in the body of the paper related to this statement.

Having established the clinical, neuroradiological, neuropsychological, and neuropathological findings that characterize chronic neurological disease in boxers, we now turn our attention to American football players.

AMERICAN FOOTBALL

In 2005, Omalu et al. [18] reported a case of the brain neuropathology of a deceased 50-year-old former National Football League (NFL) player whose NFL career spanned 17 years after playing in high school and college. The gross pathology consisted of decreased pigmentation of the substantia nigra and otherwise was normal. Routine histology reportedly revealed neuronal dropout in the substantia nigra and cerebellar cortex but this was not depicted in accompanying images or quantified in any manner. No Lewy bodies were present. “The battery of immunohistochemical stains revealed frequent diffuse extracellular amyloid plaques, sparse-positive neuritic threads, and sparse intraneuronal band-shaped and flame-shaped neurofibrillary tangles (NFTs) in the frontal, temporal, parietal, occipital, and cingulate cortex and the insula.” The authors wrote that these findings “met criteria for CTE”, specifically citing Corsellis et al.’s 1973 [12] report on the neuropathology of chronic traumatic encephalopathy (CTE) in boxers. Omalu et al. stated that “possible symptoms of CTE may include recurrent headaches, irritability, dizziness, lack of concentration, impaired memory, and mental slowing; mood disorders, explosive behavior, morbid jealousy, and pathological intoxication and paranoia, tremor, dysarthria, and parkinsonian movement disorders”. The subject’s premortem medical history (obtained from posthumous telephone interviews with relatives of the deceased) of “dysthymic” mood disorder, deficits in memory and judgment and parkinsonian symptoms was noted. There was subsequently an exchange of letters between Casson et al. and Omalu et al. discussing the validity of Omalu et al.’s statements [19, 20].

In 2005, Guskiewicz et al. [21] reported the results of a study on memory impairments and AD in retired NFL players. The authors mailed self-report health questionnaires to 3,683 retired players; 2,552 participated by returning the forms. Thirty-three (1.3%) of these indicated that they had been diagnosed with AD by a physician. There was no correlation with the number of concussions reported, but the authors stated that AD occurred in younger ages in the subjects compared to the general public. A mild cognitive impairment (MCI)/memory questionnaire was filled out by a subset of 758 retired players and in 641 cases also a spouse or other close relative. Twenty-two subjects reported that they had been diagnosed with MCI by a physician. Another 77 subjects were reported to have “significant” memory impairments according to their spouses or other close relatives. The authors reported that there was a significant correlation between the prevalence of MCI/memory impairments and subjects’ history of having sustained 3 or more concussions during their playing careers.

In 2006, Omalu et al. [22] reported “the second autopsy-confirmed case of chronic traumatic encephalopathy in a retired professional football player”. They stated that the neuropathological features differed from those in their first case; “In contrast to the first case, the brain in this second case revealed topographically distributed sparse to frequent NFT (neurofibrillary tangles) and NT (neuritic threads) in the neocortex, hippocampus, subcortical ganglia, and brainstem. There were no diffuse amyloid plaques.” There was a non-fenestrated CSP and mild pallor of the substantia nigra. Neuronal dropout was reported in the substantia nigra, the cerebellum and the brainstem. This was not quantified or depicted in accompanying images. Premortem medical history was obtained posthumously from surviving relatives and review of medical records from a psychiatric hospital. There is no mention of any objective neurological examinations. The 45-year-old subject had played football for 2 years in the military, 4 years in college, and 8 years in the NFL. He had a history of major depression, mood swings, multiple suicide attempts (cause of death was a completed suicide), multiple business failures, and being indicted for arson.

In 2007, Guskiewicz et al. [23] reported on the prevalence of depression in retired NFL players. The study data was collected from the same self-report questionnaires returned by the same group of subjects that was discussed previously [21]. Two hundred and sixty-nine of the 2,552 subjects (11%) reported that they had previously or currently been diagnosed with depression. The authors stated that this prevalence “is generally consistent with the lifetime prevalence in the general U.S. population”. They reported that the prevalence of depression was three times greater in subjects with a history of 3 or more concussions compared to subjects with a history of fewer than 3 concussions.

In 2009, McKee et al. [24] reported the neuropathological findings in 2 retired pro boxers and 1 retired pro football player. They also reviewed the 51 cases of “neuropathologically confirmed CTE” in the literature (including the 3 cases they reported in this paper). They stated that repeated closed head injuries occur in many contact sports and that in collision sports like football and boxing “players may experience thousands of subconcussive hits during the course of a single season”. The 51 cases they reviewed included 39 ex-boxers and 5 ex-professional football players. They stated that the clinical picture of “CTE” gleaned from this review began with cognitive and psychiatric/ behavioral symptoms which progressively worsened and in severe cases developed “slowing of muscular movements, abnormal gait, masked facies, vertigo, speech problems and tremors”. They (incorrectly) cited Corsellis et al. [12] as reporting 3 stages of clinical deterioration as follows:

“The first stage is characterized by affective disturbances and psychotic symptoms.

McKee et al. [23] reported that 14 of the 45 cases (boxers and football players) had a mood disturbance, usually depression, early in the course and that “movement abnormalities were eventually found in 41% ” of the cases. They cited Corsellis et al. [12] in the discussion of the neuropathological findings, listing the major findings (CSP and fenestrations of the cavum listed fourth). McKee et al.’s [24] review of the literature revealed a CSP in 69% of cases and fenestrations in 49%. Review of the graphic presentation of the neuropathology of the cases indicates that the brain of the new football player subject in this paper did not have a CSP or fenestrations but the brains of the new boxer cases presented in this paper both had CSPs and one had fenestrations of the septum as well. McKee et al. [24] emphasized the histological and immunocytochemical findings of widespread neurofibrillary changes and tau deposition in the superficial cortical layers (II and III), often in a perivascular distribution. They stated that these neuropathological findings, specifically the distribution of abnormalities, correlated with the clinical picture of behavioral and cognitive dysfunction and the later appearance of parkinsonism, ataxia and speech disturbances.

In 2010, Omalu et al. [27] reported on 5 cases (4 ex American football players and 1 wrestler) of neuropathologically confirmed CTE that had exhibited suicidality while alive. Two of the football players were previously reported by Omalu et al. in 2005 and 2006 [18, 22]. Clinical information on all 5 cases was obtained from retrospective, posthumous interviews with relatives and/or friends of the deceased. The authors reported that the 5 cases had demonstrated behavioral symptoms such as memory loss, abnormal executive functioning, paranoid ideation, depression, and suicidality. They concluded that “CTE refers to chronic cognitive and neuropsychiatric symptoms of chronic neurodegeneration following a single episode of severe traumatic brain injury or (more commonly) repeated episodes of mild traumatic brain injury”.

In 2011, Omalu et al. [26] reported on neuropathology of 14 professional athletes (including 8 retired American football players and 1 ex-boxer (some of these cases had been previously reported by Omalu et al). Seven of the 8 football players and the boxer were neuropathologically diagnosed with CTE using the following criteria: “The fundamental neuropathologic feature of CTE is the presence of sparse, moderate, or frequent band-shaped, flame shaped, small globose, and large globose NFTs in the brain accompanied by sparse, moderate, or frequent neuritic threads “in a characteristic topographic distribution including perivascular and depths of sulci locations”. Only one of the cases in this series (the 1 ex-boxer) was reported to have abnormalities of the septum (fenestrations). Unlike McKee et al.’s [24] cases, Omalu et al.’s [26] cases did not demonstrate that these neurofibrillary changes preferentially involved the superficial cortical laminae. Omalu et al. [26] also pointed out a number of other differences between the neuropathologies in this series and those described by McKee et al. in 2009 [24]. Omalu et al. [26] suggested that these differences might be related to the fact that most of the cases of CTE reported by McKee et al. [24] in that literature review and case report were boxers whereas most of the cases Omalu et al. were presenting were in football players with only one being a boxer. “The differences between their conclusions and ours may suggest that there may be subtle neuropathologic distinctions between CTE found in different sports professionals.” The clinical features were similar to those reported in Omalu et al.’s 2010 report [27].

In 2013, McKee et al. [28] published the neuropathological findings of “CTE” in the brains of 85 athletes, 34 of whom were retired American football players. In the Introduction, the authors linked the paper to work on dementia pugilistica in boxers beginning with Martland in 1928 [1] and repeated their earlier statements [24] regarding the predominance of symptoms of behavioral/psychiatric (including suicidality) and cognitive dysfunction in cases of CTE dating back to those earlier reports in boxers. Medical histories were obtained posthumously from families and friends of the deceased. CTE was diagnosed by histological/immunocytochemistry findings of specific patterns of hyperphosphorylated tau in these brains. The authors divided up the cases into 4 stages of CTE based on extent of tau deposition, with stage 1 being the mildest and stage 4 the most severe. None of the 7 stage 1 cases had septal abnormalities, 4 of the 14 stage 2 cases had small CSPs, 5 of the 15 stage 3 cases had septal abnormalities, and “most” of the 15 stage 4 cases had large CSPs or septal perforations or absence of the septum. The authors reported the following clinical classification:

Stage I chronic traumatic encephalopathy included headache and loss of attention and concentration;

The authors specifically noted that 13 of the 51 CTE cases in this series had suicidality (7 completed suicides and 6 with suicidal ideation). The authors also stated that the neuropathological stages of CTE correlated with number of years of football played. Finally, McKee et al. [28] stated that there was a clinicopathological correlation between the neuroanatomical areas “affected by CTE” and the behavioral/psychiatric and cognitive dysfunctions that make up the clinical picture.

In 2013, Hart et al. [29] reported the results of neurological and neuropsychological testing in 34 retired NFL players (ages 47–71) and the results of MRI imaging in 26 of these same subjects. The results were compared to those of matched controls. Results of routine anatomical MRI brain imaging were not reported; the results of advanced 3T MRI techniques including FLAIR, diffusion tensor imaging, hemosiderin scanning and arterial spin labeling were reported. On NP testing, 14 subjects (41%) demonstrated” cognitive deficits”: 4 had “fixed” deficits, 8 had MCI, and 2 had dementia. Eight of the 34 (24%) were diagnosed as having depression. Overall, the NP testing revealed “mild difficulties” in naming, word finding, and episodic verbal and visual memory. Detailed results of various NP subtests are included in Table 1. The results of the clinical neurological evaluations of all the subjects are not explicitly stated. The report indicates that 7 of the NFL players complained of headaches and 1 complained of dizziness, that “no neurological abnormalities suggestive of” amyotrophic lateral sclerosis were detected and “that none of the retired players fit the reported clinical profile for CTE”. The authors reported that cognitive impairments and depression in the NFL players correlated with changes in blood flow to specific brain regions and white matter abnormalities detected on advanced MRI imaging.

In 2014, Casson et al. [30] reported the results of neurological, neuropsychological, and MRI testing of 45 retired NFL players between the ages of 30 and 60. The subjects were recruited from a list of over 5,000 retired NFL players supplied by the players’ union. The testing procedures were modeled after the author’s 1984 study of living retired boxers [16]. Neurological examinations revealed isolated Babinski signs in 3 subjects, mild sustention-intention tremors in 2 subjects, horizontal nystagmus in 1 subject, and abnormal dynamic visual acuity testing in 3 subjects. There were no instances of dementia, dysarthria, cerebellar or extrapyramidal signs. On neuropsychological testing, there were no cases of dementia, 28 subjects (62%) were normal, 11 subjects (24%) had impairments on 1 or 2 subtests (possible MCI), and 6 subjects (13%) had borderline to mild impairments on 1 or 2 subtests confounded by low verbal IQ or lack of effort (Table 1). Anatomical MRI scans revealed 3/45 subjects with a large CSP and cerebral atrophy. Thirty-one subjects had small CSPs which the authors pointed out was similar to the prevalence of small CSPs on MRI in the general population. Advanced MRI techniques revealed 4 subjects with microbleeds on susceptibility weighted imaging and mean fractional anisotropy results on diffusion tensor imaging for the entire group that correlated with number of concussions sustained in the NFL and number of years of pre-high school and high school football played. The authors concluded that “MRI lesions and neuropsychological impairments were found in some players; however, the majority of retired NFL players had no clinical signs of chronic brain damage.”

In 2016, McKee et al. [31] reported on the results of a NINDS-NIBIB consensus meeting on the neuropathological criteria for the diagnosis of CTE. “The panel defined the pathognomonic lesion of CTE as an accumulation of abnormal hyperphosphorylated tau (p-tau) in neurons and astroglia distributed around small blood vessels at the depths of cortical sulci and in an irregular pattern.” Other tau related pathologies and gross pathologies such as septal abnormalities were supportive of the diagnosis but not necessary for the diagnosis. The authors stated that “previous data has shown that the pathological features of CTE associated with boxing (often referred to as “dementia pugilistica”) are similar to the pathological features of CTE associated with football.” No clinical criteria for the diagnosis of CTE were mentioned.

In 2017, Mez et al. [32] reported on the neuropathological findings in 202 American football players, 111 of whom had played professionally in the NFL. Using the NINDS-NIBIB consensus refined neuropathological criteria for the diagnosis of CTE (i.e., at least 1 perivascular phosphorylated tau (p-tau) deposit at the depths of cortical sulci), 110 of the 111 NFL players were diagnosed with CTE. Stages 1 and 2 CTE cases were considered mild; stages 3 and 4 CTE cases were considered severe. Clinical information on the cases was obtained retrospectively (posthumously) from family and friends of the deceased. The authors stated that there were 2 general clinical presentations of CTE: 1) behavior and/or mood symptoms in those presenting at a younger age, and 2) cognitive symptoms in those presenting at an older age. Among the behavior/mood symptoms frequently seen in these cases were impulsivity, depressive symptoms, apathy, and anxiety; less commonly reported were hopelessness, explosivity, verbally or physical violent, and suicidality. Dementia and motor symptoms were common in severe cases. Among the 27 mild CTE NFL cases, 26 had behavioral and/or mood symptoms and 23 had cognitive symptoms. Among the 84 severe CTE NFL cases, 75 had behavioral and/or mood symptoms and 80 had cognitive symptoms. Despite claiming that in severe cases of CTE the anatomical distribution of p-tau correlated with behavioral/mood/cognitive clinical symptomatology, the authors also noted that “participants with mild CTE pathology often had these symptoms despite having relatively circumscribed cortical pathology and absence of ptau pathology in the hippocampus, entorhinal cortex or amygdala.”

COMPARE AND CONTRAST THE CHRONIC NEUROLOGICAL CONSEQUENCES OF REPETITIVE HEAD INJURIES IN BOXING AND AMERICAN FOOTBALL

Although the initial reports regarding both boxers and football players were authored by forensic pathologists [1, 18], the means of investigation and research methods involved in evaluating athletes in the two sports diverged immediately following these bylines. In the 40 years following Martland’s report, well over 100 cases of the clinical neurological findings in living retired boxers were described in the literature [1, 9]. There were detailed descriptions of neurological signs in individual cases as well as descriptions of the neurological syndromes seen in large groups of retired boxers that had been personally examined by authors. Then, in 1969, AH Roberts [2] published his classic monograph in which he described the unique clinical pattern of neurological findings that defined a chronic disease of ex-boxers, based upon his personal examinations of 224 randomly selected living retired boxers (37 of the 224 exhibited signs of this disease). Although a few autopsy reports of brains of boxers had appeared over the years, it was not until Payne [11] and then Corsellis et al. [12], that a definitive, unique pattern of neuropathological abnormalities in retired boxers was described. Payne [11] reported on the neuropathological findings in the brains of subjects whom he had previously examined in life. Corsellis et al.’s [12] neuropathology reports were accompanied by clinical information obtained from posthumous interviews with relatives of the deceased as well as reviews of medical records. It is thus clear that the development of ideas regarding chronic CNS disease in retired boxers was, from the beginning, clinically driven and based largely upon neurological evaluations of living subjects.

This stands in sharp contrast to the neuropathologically driven origins and subsequent investigations into chronic CNS neurological disease in retired American football players. Both of the initial papers by Omalu et al. [18, 22] reported on a single case of the neuropathological findings in a deceased former NFL player. The only clinical information came from posthumous statements by relatives of the deceased. No objective reports of neurological examinations during life were included. Later reports of larger series of neuropathology findings in deceased NFL players by Omalu et al. [26, 27] followed this same pattern of clinical information collection. When McKee et al. [24] started reporting on neuropathology in retired football players, they included a review of the literature of “neuropathologically verified cases” of chronic brain disease related to repetitive brain trauma (mostly ex-boxers) without reviewing the large number of clinically defined cases (without neuropathology) that make up the bulk of the cases in the boxing literature. McKee et al.’s later reports [28, 32] of large numbers of cases of “CTE” in retired NFL players are based on neuropathological findings in the brains of deceased subjects with clinical information obtained from more structured interviews with relatives and friends of the deceased but still without objective neurological examination findings while the subjects were alive.

In addition to these neuropathology reports, two reports of data on depression and cognition from a large number of retired NFL players have been reported; both relied solely on self-report questionnaires from the subjects and their spouses and neither include any objective neurological examinations [21, 23]. The two studies of clinical examinations of living retired NFL players in the literature found some changes on advanced MRI, some cognitive/memory dysfunction, a prevalence of depression similar to or slightly higher than the general population, and essentially unremarkable clinical neurological examinations [29, 30]—none of which seemed to correlate with the neuropathology that has been reported by Omalu et al. [18, 25–27] and McKee et al. [24, 28, 32].

In view of these divergent approaches to investigating chronic CNS disease, it should come as no surprise that the clinically-based research into boxers should result in a clinically-defined syndrome which correlates with a well-defined neuropathological picture and that the neuropathologically-based and centered research into football players has led to a purely neuropathologically-defined condition.

There are clear differences between the descriptions of the long-term clinical effects on the brain in boxers versus football players. The clinical picture in football players is predominantly mood/behavioral and memory/cognitive in nature. “Motor symptoms” including parkinsonism and gait disturbances are described as occurring late in the disease and are mentioned mostly in passing. The reports from Omalu et al. [25–27] emphasized behavioral/mood abnormalities such as suicidality, depression, and paranoid ideation along with abnormalities of memory and executive functions. In their 2009 paper, McKee et al. [24] stated that the clinical picture of “CTE” gleaned from their review began with cognitive and psychiatric/behavioral symptoms which progressively worsened and in severe cases developed “slowing of muscular movements, abnormal gait, masked facies, vertigo, speech problems and tremors”. In McKee et al.’s [28] study on a large series of cases (mostly football players), McKee et al. [28] described clinical features of mild cognitive impairments, explosive behaviors, and depression in early, milder cases progressing to dementia, executive function dysfunction, and aggressive behavior in later stages. In Mez et al.’s [32] paper reporting a large series of cases of “CTE” in football players, they stated that there were two general clinical presentations of CTE: 1) behavior and/or mood symptoms in those presenting at a younger age; and 2) cognitive symptoms in those presenting at an older age.

Among the behavior/mood symptoms frequently seen in these cases were impulsivity, depressive symptoms, apathy, and anxiety; less commonly reported were hopelessness, explosivity, verbally or physical violent, and suicidality. Dementia and motor symptoms were common in severe cases.

In their reports, Omalu et al. [18, 25–27], McKee et al. [24, 31], and Mez et al. [32] often stated that the clinical symptoms they were reporting in American football players were consistent with, similar to or the same as those reported by the earlier studies of retired boxers. The review of the boxing literature that is the focus of this present review clearly indicates that such statements are not correct. It is possible that a methodological error may be, in part, to blame for these different points of view. In McKee et al.’s [24] review of the boxing literature, only “neuropathologically verified cases of CTE” were included. McKee et al. [24] thus inadvertently ignored the purely clinical reports of Winterstein [3], Critchley [4] (his study was referenced in the footnotes only to support the concept that the condition is progressive), AH Roberts [2], and others that laid out in great detail the clinical neurological descriptions that characterize the disease in living ex-boxers.

Beginning with Martland [1] and continuing along through Winterstein [3], Critchley [4], Spillane [8], Mawdsley and Ferguson [9], and then AH Roberts [2], the descriptions of the clinical features in boxers emphasize the neurological signs of dysarthria, cerebellar dysfunction, extrapyramidal dysfunction (parkinsonism), and pyramidal dysfunction, occurring in varying combinations. These authors emphasize the unique nature of these neurological patterns seen in boxers as being distinct (but reminiscent) from those seen in other neurological conditions. Memory impairments and dementia were seen in some cases (16 of 37 in AH Roberts’ monograph [2]) but generally in the more severe cases and almost always accompanied by the abnormal neurological signs previously mentioned. For example, only 1 of the 224 ex-boxers examined by AH Roberts [2] had significant memory deficits unaccompanied by the abnormal neurological signs of chronic brain damage. Psychiatric/behavioral disorders were present in some cases, again almost always accompanied by abnormalities on the clinical neurological examination. Depression was mentioned in only a small number of cases. Suicidality was rarely reported. When Johnson [10] reported on psychiatric disorders in 16 retired boxers, he noted that in all 16 the neurological signs of disease appeared prior to the psychiatric impairments. Johnson [10] reported 4 types of psychiatric disorders in these retired boxers: amnestic syndrome, morbid jealousy, psychosis, and personality disorder with rage reactions/impulsive aggressive behaviors. There was no mention of suicidality or depression. In summary, the clinical features reported in football players are predominantly mood/behavior/cognitive in nature whereas those reported in boxers are predominantly neurological in nature; when cognitive/behavioral symptoms occur in boxers, they usually are seen late in the course, concomitantly with abnormal neurological signs and are of different types than have been reported in football players.

There are a number of possible explanations for these major differences in clinical features. A likely reason is that the chronic CNS disease of boxers is a distinctly different condition than the chronic CNS disease of American football players. Another possibility is that the different methods of ascertainment of the clinical features in the subjects (neurological examinations of living retired boxers versus posthumous collection of medical history from the relatives and friends of deceased football players) has resulted in widely differing information being collected. It is also possible that the selection bias in the football cases (i.e., families of deceased players who manifested behavioral/mood/cognitive problems in life may be more likely to donate their loved ones’ brains to brain banks than families whose loved ones manifested only physical or no symptoms in life), may have affected the reported clinical picture. This selection bias could also result in the inclusion of some clinical symptoms that are unrelated to the effects of head trauma.

There are also a number of significant differences between the brain neuropathology reported in football players and that reported in boxers. The unique tetrad of pathology in boxers reported by Corsellis et al. [12] (septal abnormalities, cerebellar scarring and neuronal loss, gross depigmentation of the substantia nigra with near total loss of pigmented neurons and regional occurrence of neurofibrillary tangles diffusely throughout the brain with scarce or no senile plaques – amended in 1990 by GW Roberts [17] to include diffuse immunoreactivity to amyloid proteins) is not what has been reported by Omalu et al. [18, 25–27] and later McKee et al. [24, 31] in football players. The latter have focused their descriptions on the histopathology of neurofibrillary changes/p-tau deposition in specific topographic distributions in the brain. Septal abnormalities which were present in all 15 of Corsellis et al.’s [12] boxing cases and all 6 of Payne’s [11] boxing cases have not been present in many of the “mild” football cases but are more common in the “severe” football cases. As McKee et al. [31] have refined the pathological definitions of chronic brain damage in football players, the presence of septal abnormalities and other gross abnormalities has been more and more de-emphasized. According to present NINDS-NIBIB criteria [31], the sole feature required for a diagnosis of “CTE” in football players is the presence of 1 or more foci of perivascular accumulations of p-tau in neurons and astroglia in the depths of sulci in an irregular pattern. Despite Omalu et al.’s and McKee et al.’s claims to the contrary, this is a far cry from Corsellis et al.’s description [12]. In a 2018 study of immunostaining of material from 14 of Corsellis et al.’s 15 boxing cases, Goldfinger et al. [33] found that only 7 of the 14 met the NINDS-NIBIB pathognomonic criteria for the diagnosis of “CTE” in football players; this serves as further evidence of the significant differences between the neuropathology of boxers and that of football players.

Comparison of the results of ancillary tests (e.g., neuroimaging studies, standardized neuropsychological testing) in living retired boxers [13, 17] and football players [29, 34] can also shed light upon the long-term effects of participation in these sports on the brain (Table 1). On the Trails making test, which is a measure of executive functioning, 22 of 28 boxers were impaired whereas only 5 of 67 football players were impaired. In another group of retired football players [34] where individual scores were not recorded, the group of 72 NFL players performed similarly on the Trails A and B tests to a group of controls. On Wechsler Memory Scale testing of verbal memory, 13 of 13 boxers were impaired and on testing of visual memory, 11 of 13 boxers were impaired. Although it is not the same exact test, football players were administered the California Verbal Learning Test of verbal memory. In one study, 12 of 45 football players were impaired on the California Verbal Learning Test: in the other study, 10 of 22 football players had mean T scores in the low normal range but it is not clear how many were impaired. Although it is not exactly the same test as the Wechsler visual memory test, football players were administered the Brief Visuospatial Memory Tes test of visual memory. Eight of 45 football players were impaired on the Brief Visuospatial Memory Tes. The total numbers of retired boxers and football players whose equivalent standardized test results are available for comparison is small. The results do suggest that retired boxers exhibit a much higher prevalence of impaired function in the spheres of executive function and verbal/visual memory compared to retired football players.

Comparison of neuroimaging studies between boxers and football players is not straightforward, mainly because different neuroimaging studies were used in different eras. Thus, PEG studies were performed on boxers in the 1960 s [7, 35], CT scans were performed on boxers in the 1980 s and 1990 s [13–15, 37], and MRI scans were performed on football players [29, 38] (and in one study on active boxers [39]) from 2000 onwards. With this in mind, Table 2 summarizes the neuroimaging findings in the literature. It is important to note that MRI scanning is much more sensitive in visualizing anatomical structures such as the septum pellucidum than CT or PEG. Therefore, the prevalence of CSPs on MRIs may not be comparable to the prevalence on CTs due to better technology. Early generation CT scans were unlikely to detect small CSPs, so any that were visualized on those CT scans were probably large. Review of the studies summarized in Table 2 suggests that cerebral atrophy was more commonly seen on neuroimaging of retired boxers than of retired football players; however, since only one of the MRI studies of retired football players actually assessed cerebral atrophy [30], this awaits further investigations. Large CSPs on neuroimaging seem to be more common in both groups than in the general population. MRIs and CT scans have demonstrated that CSPs can develop over time in active boxers [36, 40]. It certainly appears that repetitive brain trauma is linked to large CSPs.

Based upon this review of the medical literature, Table 3 compares the main clinical and neuropathological features of the chronic CNS conditions in retired boxers and retired American football players [1–39].

BIOMECHANICS OF BOXING AND AMERICAN FOOTBALL HEAD IMPACTS

Differences between the head impacts sustained in football and boxing may partially explain the long-term neurological consequences of repetitive head injuries in players of the two sports. Football players wear hard shell helmets and faceguards during practice and games while professional boxers wear no headgear during bouts (they may wear soft headgear while sparring). Boxers frequently sustain multiple blows to the head in rapid succession often with only minimal intervals between impacts. Football players sustain head impacts in rapid succession much less frequently due to the intervals between plays that are part of the sport. The differing biomechanics of head impacts between the two sports, although less obvious to the general public, may play an even greater role.

The biomechanics of punches has been studied using surrogates for the opponent. Atha et al. [40] used a single boxer and an instrumented ballistic pendulum to evaluate a single straight punch. The professional boxer punched the surrogate with 4,096 N, which the author estimated translated into 6,320 N of force to a human head. This force produced peak acceleration for 53 g on the 7 kg ballistic pendulum. Joch et al. [41] placed 70 boxers into one of three categories, including 24 elite, 23 national, and 23 intermediate boxers. The force of straight right punches was measured with a water-filled punching bag fit and pressure transducer. The maximum punch force ranged from 2,932 N – 3,453 N.

Smith et al. [42] developed a boxing dynamometer to measure punch force. Twenty-three boxers were sorted into elite, intermediate, and novice boxer categories. The elite boxers had a mean rear-hand punch force of 4,800 N and a front-hand punch force of 2,847 N. The intermediate boxers’ rear and front hand punch forces were 3,722 N and 2,283 N, respectively, and the novice boxers’ mean rear and front hand punch forces were 2,381 N and 1,604 N.

Walilko et al. [43] studied the biomechanics of straight punches to the jaw causing translational and rotational head acceleration. Seven Olympic boxers from five weight classes delivered 18 straight punches to the compliant face of the Hybrid III dummy. The punch force averaged 3,427±811 N, hand velocity 9.14±2.06 m/s and an effective punch mass 2.9±2.0 kg. The jaw load was 876±288 N. The peak translational acceleration was 58±13 g, rotational acceleration 6,343±1789 r/s², and neck shear 994±318 N. The severity of the punch increased with weight class primarily due to a greater effective mass of the punch.

Viano et al. [44] reported on the punch biomechanics of 11 Olympic boxers weighing 51 kg (112 lb) to 130 kg (285 lb). The effective mass of the arm was measured and the tests followed the methods of Walilko et al. [43]. A Hybrid III dummy with a frangible face was used to represent the response of the jaw and realistically transfer acceleration to the head. For the tests, a cork insert was used to give facial compliance for the straight jaw punches. There were straight punches to the forehead, hooks to the temple, and uppercuts jaw of the Hybrid III dummy.

Table 4 summarizes the Olympic boxer’s height, weight, and effective punch mass of the hand. The hand mass increased with boxer weight and averaged 1.67±0.28 kg. The average acceleration of the hand in the punch was 180.9±98.8 g with an average hand force of 2,994±1,875 N. The hook produced the greatest impact force of 4,405±2,318 N), inertial load on the head of 3,107±1,404) and neck load of 855±537 N. It also had the highest velocity change of the hand at 11.0±3.4 m/s. The hook also produced the largest head translational and rotational accelerations at 71.2±32.2 g and 9,306±4,485 r/s². The lowest forces occurred with the uppercut to the jaw. The punches resulted in 2.98±0.9 = 87 change in velocity of the Hybrid III head with an average 48.9±28.4 g head acceleration and 6,319±3,739 r/s² for all punches.

Pellman et al. [45, 46] reported on the reconstruction of NFL game impacts causing concussion. The biomechanics of the helmet impacts were determined using Hybrid III dummies in laboratory testing. Viano et al. [47] reported on the biomechanics of the struck player experiencing concussion. Table 4 summarizes the results of the NFL reconstructions. The effective mass of the striking player was estimated at 14.0 kg. This is 8.4-times greater mass than the mass of the hand in a boxing punch. Since the speeds of impact are similar, the football impacts involve more momentum in the impact. The translational accelerations of the head are greater in football impacts with concussion than in boxing (94.3±27.5 g versus 48.9±28.4 g) and resulted in higher Head Injury Criterion (HIC, 357±184 versus 52±53). This is a result of the smaller effective mass of the hand in boxing. Boxing punches produced about the same rotational accelerations of the head (6,319±3,739 g versus 6,654±1,745 r/s²). The boxer’s punch resulted in lower rotational velocity of the head than in NFL concussion (22.9±7.2 r/s versus 40.6±9.9 r/s) because of the lower momentum in the punch.

The force from the boxer’s hook exceeded that of the non-injured NFL players and was within the statistical range for concussion. The jaw and forehead impact forces were lower and the uppercut produced the lowest inertial loads on the Hybrid III head. Boxers deliver the same or more rotational acceleration to the Hybrid III dummy head for the hook and jaw punches than occurred in NFL concussions. However, the duration of impact is shorter for the boxing punch, so the rotational velocity of the head is similar to that in NFL concussion impacts with longer duration but lower rotational acceleration. The NFL reconstruction data includes the correction for problems with the rotational acceleration package used in the testing [48].

The boxers’ punches resulted in lower translational accelerations in the struck head, as compared to the football impacts. The boxers’ punches applied a higher moment to the struck head than did the football impacts. This necessarily resulted in higher rotational accelerations in the head struck by the boxers’ punch. Boxers sustain brain injury by two mechanisms, translational and rotational accelerations of the brain, with a preponderance of the rotational component. Professional football players sustained MTBI mostly by translational accelerations.

The super-heavy weight boxer generated punch forces of 5,352±2,775 N, which is above the average impact force of 7,642±2,259 N with concussion in NFL players. Since a majority of NFL players are injured by facemask or lateral impacts on the helmet, the loading direction is consistent with lateral direction of the boxing punches. Straight anteroposterior impacts are an uncommon cause for NFL concussions.

There are probably two primary means by which boxers deliver concussive blows. The first means involves the boxer delivering enough translational acceleration. The hook involves a blow to the temple, which is just above the head center of gravity (cg). The forehead punch delivers force frontally above the head cg and the jaw impact applies force below the head cg. These impacts translate the head, and the forces can reach levels consistent with NFL concussions. The damaging mechanism is translational acceleration where the greater the mass of the punch, the greater the HIC and translational acceleration. The second means involves rotational acceleration, which occurs with the impacts taking advantage of the offset from the head cg. During the punch, the axis of impact moves away from the head cg and introduces proportionately more rotational acceleration during the punch. The hook, for example, is always thrown with the elbow bent [49]. This necessarily results in the axis of impact moving away from the cg after impact, thus imparting a significant amount of rotational acceleration to the opponent’s head.

With the use of football helmets, the striking player must line up his impacts closely with the head cg of the other player. This allows the impact to transfer energy. If the impact vector is at an angle, the blow will glance off due to the smooth plastic shell of the helmets. Players need to align their impact through the head cg to deliver a solid blow and maximize energy transfer to the other player. Severe helmet impacts causing concussion involve high translational acceleration and change in head velocity (delta V). NFL concussions involve an average impact velocity of 9.3±1.9 m/s; and, the delta V is 7.08±1.8 m/s for the concussed player. Since the duration of impact is nominally 15 ms, the peak head acceleration is high at 98±28 g. In football, there is a strong correlation between translational and rotational acceleration due to the impact alignment and subsequent head-helmet motion.

In boxing, the punch and glove conform more to the head of the opponent allowing punches to induce high rotational acceleration without high translational acceleration. The effective mass of the boxer’s fist is 1.67±0.28 kg, which is more than an order of magnitude lower than the 14 kg effective mass of the helmeted football player who strikes an opponent [47]. With concussion, the striking player lines up their head, neck, and torso so their effective mass is considerable, and only the head and part of the opponent’s neck resist the blow. In boxing, the most efficient energy transfer involves more rotational acceleration than translational acceleration.

The punch velocity of the boxers averaged 6.7–11.0 m/s for the four different punches. These levels are essentially similar to the impact speed in football concussions; but, the head delta V after a punch was only 2.8–3.1 m/s on average, well below half that of the head delta V with NFL concussion. This reflects the much lower effective mass of the punch. Boxers cannot deliver high translational acceleration and delta V to the opponent because of the low punch mass in comparison. Boxers deliver rotational accelerations in and above the range where NFL players are concussed.

These significant differences between the biomechanical forces impacting the head and brain in football as opposed to boxing are likely a partial explanation for the differing long-term neurological effects seen as the result of repetitive head injuries in the two sports.

CONCLUSION

Participants in boxing and American football are often subject to multiple repetitive head injuries over time. This review of the medical literature has focused on the possible long-term effects of such exposure on the CNS. The evidence indicates that there are significant differences between the clinical neurological features of chronic brain damage seen in boxers and those seen in football players. There are also significant differences in the brain neuropathology that has been described in boxers as compared to that described in football players. The clinically driven approach to studying boxers that is evidenced in reviewing the boxing literature has resulted in a well-defined clinical picture and a well-defined neuropathological picture that correlates well with that clinical picture. The pathology driven approach to studying football players has resulted in a highly refined and specific neuropathological definition of chronic brain damage in football players with uncertain correlations to a clinical picture mostly arising from the highly subjective retrospective recollections of relatives of deceased players rather than from objective clinical neurological examinations of living retired players. Differences in the biomechanical impact forces in boxing and football may be a part of the explanation for the differences between the chronic CNS disease of boxers versus football players.