The Physician and Sportsmedicine
Menubar Home Journal Personal Health Resource Center CME Advertiser Services About Us


Acute Traumatic Brain Injury in Amateur Boxing

Erik J. T. Matser, PhD; Alphons G. H. Kessels, MD; Muriel D. Lezak, PhD; Jaap Troost, MD; Barry D. Jordan, MD


BACKGROUND: Acute traumatic brain injury (ATBI) represents the neurologic consequence of concussive and subconcussive blows to the head. Evidence suggests that ATBI may be associated with boxing and collision sports such as American football and soccer, thus potentially exposing millions of athletes annually.

OBJECTIVE: The objectives of this study were to determine whether significant ATBI occurs in boxers who compete and, if present, the nature of the cognitive impairment. A secondary objective was to determine if headgear could reduce the risk for ATBI in amateur boxing.

DESIGN: In this inception cohort study, 38 amateur boxers underwent neuropsychological examination before and shortly after a boxing match and were compared with a control group of 28 amateur boxers who were tested before and after a comparable physical test. The main outcome measures were neuropsychological tests (memory, mental and fine-motor speed, planning, and attention) proven to be sensitive to cognitive changes incurred in contact and collision sports.

RESULTS: The boxers who competed exhibited an ATBI pattern of impaired performance in planning, attention, and memory capacity when compared with controls. They had significantly different findings in the Categorization Task Test (P= 0.047); Digit Symbol Test (P= 0.02); Logical Memory: Short Term Memory and Long Term Memory subtests (both tests, P< 0.001); and Visual Reproduction: Short Term Memory subtest (P< 0.001) and Long Term Memory subtest (P< 0.03).

CONCLUSION: Participation in amateur boxing matches may diminish neurocognitive functioning despite the use of headgear. The neurocognitive impairment resembles cognitive symptoms due to concussions. Guidelines are needed to reduce the risk for repeated ATBI.

Moderate-to-high incidences of acute traumatic brain injury (ATBI) are reported among participants in many common sports such as American football, soccer, basketball, baseball, rugby, martial arts, and ice hockey (1-4). ATBI incurred in contact and collision sports can result in functional alterations that range from transient cognitive impairments to death (5). Commonly, evidence of ATBI in the athlete appears as temporary cognitive function deficit, loss of consciousness, amnesia, postconcussion syndrome, subdural hematomas, headaches, or convulsions (5-9).

Of all sports, boxing is associated with a large number of deaths relative to the number of participants and ATBI (7). Because the objective of a boxing match is to make one's opponent unable to fight, it is not surprising that ATBI occurs in boxing matches and sparring sessions. Jabs and angled blows to the head deliver acceleration-deceleration, linear, and rotatory energy vectors that may result in ATBI.

In addition to ATBI, chronic traumatic brain injury (CTBI) is of concern in boxing. CTBI from sports activity is often insidious. Onset begins with mild subclinical dysfunction detected only through neuropsychological investigation (10-12) and evolves into the slowed motor performance, tremors, cognitive deficits, and personality and psychological changes typical of chronic traumatic boxer's encephalopathy (punch drunk syndrome) (13-17). The amount and severity of ATBIs received during a boxing career appear directly related to the severity of CTBI (5). This conforms to the cumulative effect model, which states that repeated blows to the head have an additive deleterious effect (18,19).

The current investigation was undertaken to better understand ATBI symptoms and the effects of ATBI on cognitive impairment in Dutch boxers.


Subjects. The inception cohort consisted of 38 male amateur boxers and 28 male controls. The boxers who competed ("competitors") were selected randomly at fight nights. One boxer was selected at random from every bout. The control boxers ("controls") were selected from four licensed Dutch boxing teams and matched according to frequency of boxing, boxing level, weight, age, and education. All participants were Dutch nationals and educated in The Netherlands. Participation criteria excluded those who had a learning disorder, consumed more than five alcoholic drinks a week, had concussions not related to boxing, used recreational or hard drugs, or had medical conditions that might affect cognitive functioning (reported by their physicians).

Data on age, weight, boxing skill, and education were obtained in interviews. Boxers were classified into three weight categories: light (86 to 139 lb [40 to 64 kg]), middle (140 to 161 lb [65 to 75 kg]), and heavy (162 lb or over [more than 75 kg]). Education levels were classified according to the Dutch secondary schooling system: LBO (lower, <10 years), MBO (middle, 10 years), HBO (higher, >=11 years). Boxing skill was ranked according to the system used by the Dutch Boxing Federation (N, C, B, and A). A-ranked boxers are the most experienced fighters and have won more than 12 matches, N-ranked boxers have won fewer than 3 matches. The number of punches to the head was determined by reviewing the score cards of the ringside referees and then averaged. All the competitors used protective headgear during their fights to comply with Dutch Boxing Federation regulations.

Boxing statistics. Comparisons were made between the competitors and controls. Competitors were given neuropsychological tests before and after a boxing match, and controls were tested before and after they performed a bag-punching task (three times for 3 minutes each session). Documented trauma index variables included the number of previous bouts, the time elapsed before testing after a match, impact (weight class plus the total number of head punches in the match), match results (win, draw, or loss), ranking, and special circumstances (knockouts or technical knockouts). In this study, technical knockouts were scored when boxers showed symptoms of distress after being hit in the head and the fights were stopped by the referee. Fights stopped for other distress—for example, from body blows, punches to the liver, or eyebrow cuts—were not classified as technical knockouts.

Neuropsychological testing. The neuropsychological tests given have been shown to be sensitive for detecting cognitive impairment from contact and collision sports (4,11,19). Each participant (competitor or control) received the same battery of neuropsychological tests. Tests were administered by a trained psychometrician according to standardized instructions and procedures. Responses were scored by a registered neuropsychologist who was blinded to participant status (competitor or control).

The test battery included the Categorization Test (20), Paced Auditory Serial Addition Task (PASAT) (21), Digit Symbol (Wechsler Adult Intelligence Scale) (22), Trailmaking Test (A and B) (23), Stroop Test (interference trial) (24), Wechsler Memory Scale (WMS) subtests: Visual Reproduction and Logical Memory (25), and the Puncture Test (26). The Categorization Test is a word learning test that contains nine different elements that can be classified into three groups (fruit, vegetables, and trees). The test can be divided in two different cognitive functions: (1) word learning and (2) planning/organization. The WMS subtests were administered with an immediate recall and a 20-minute-delayed recall. Only one of the two subtests of the WMS Logical Memory Test was administered; for the visual memory part, we selected the most complex form of the WMS Visual Reproduction Test. The competitors and boxer controls used the same version of these tests before and after competition or the physical task.

Statistical analysis. Competitors' and controls' before- and after-exertion scores were measured and mean changes in neuropsychological test scores were calculated. Differences between groups were assessed with the student's t-test. Differences between the means, confidence intervals, and one-tailed probabilities (P-values) were also determined. Linear regression analysis was done to compare groups, with adjustments for age, boxing skill, level of education, weight, and time interval after exercise. To determine whether boxing-related variables and changes in test scores were associated among competing boxers, adjusted regression coefficients and their P-values were calculated using a multivariate linear regression model that corrected for the variables. Statistics were calculated by using a computer program (Statistical Package for the Social Sciences).


Population characteristics. Comparisons between the population characteristics of the two groups are shown in table 1. No significant differences were found in any of the parameters measured. Level of education (P=0.67) and age (P=0.37) of the competitors were not different from that of the controls. The distribution of boxers in weight classes and their skill levels were also similar. The only variable that came close to statistical significance among the population parameters was the time that had elapsed after exertion before testing (P=0.053); competitors' time was slightly longer.

TABLE 1. Comparison of Boxing Competitor and Control Group Parameters in a Study of Traumatic Brain Injury

Parameter Competitors (n=36) Controls (n=28) P-value*

Median age (range) 22 (18-30) 22 (16-30) 0.37

Education level (% low, middle, high) 40, 42, 18 41, 30, 30 0.67

Weight category (% light, middle, heavy) 47, 32, 21 44, 33, 22 0.83

Boxing skill (% N, C, B, A) 24, 26, 0, 50 26, 19, 15, 41 0.75

Median time between exertion and testing, in min (range) 5 (1-20) 4 (1-15) 0.053

Probabilities from Mann-Whitney test. P-values greater than 0.05 are not statistically significant.

Head trauma. During these amateur boxing matches, competitors sustained a median of 8 (range, 0 to 31) punches to the head. Sixty-five percent of the competitors received 10 or fewer head punches, and 35% incurred more than 10. Concussions with or without loss of consciousness were frequent. Thirteen percent of the fights ended in a knockout (concussive state with loss of consciousness) or a technical knockout (concussive state without loss of consciousness; for example, when the competitor was unable to continue because of mental impairment).

Neuropsychological testing. Table 2 shows the means of the test scores and their unadjusted differences. All the tests were also analyzed using a multivariate linear regression model adjusting for time after match or physical task, level of education, weight category, and level of boxing skill. The results differed only marginally. Competitors had more cognitive impairment in tests of planning, attention, and memory than did controls. Significant differences were found in several tests including the Categorization Task Test (P=0.047), Digit Symbol (P=0.02), Logical Memory Short Term Memory, Logical Memory Long Term Memory (P<0.001), Visual Reproduction Short Term Memory (P<0.001), and Visual Reproduction Long Term Memory (P=0.003).

TABLE 2. Mean Neuropsychological Test Score Changes After a Boxing Match (Competitors) or Punching Bag Exertion (Controls); Between-Group Differences, 90% Confidence Interval (CI)

Score Change
Neuropsychological Test Competitors Controls Between-Group Difference (90% CI) P-value*

Categorization Task
   Numbers remembered -0.05 0 -0.05 (-0.67 to 0.57)
   Number of categories -0.08 +0.19 -0.27 (-0.54 to -0.003) 0.047

PASAT -1.22 -1.17 -0.06 (-2.27 to 2.16)

Puncture test
   Dominant hand +3.00 +5.31 -2.31 (-6.41 to 1.80)
   Nondominant hand +5.24 +6.76 -1.52 (-6.38 to 1.76)

Stroop Interference Card
   Number of seconds -7.16 -7.65 +0.91 (-6.48 to 7.46)
   Number of mistakes -0.24 -0.62 +0.43 (-0.42 to 1.17)

Digit Symbol +5.66 +9.38 -3.73 (-6.63 to -0.82) 0.02

Trailmaking Test
   Form A -4.42 -3.63 -0.79 (-3.24 to 1.66)
   Form B -9.45 -11.27 +1.82 (-3.59 to 7.23)

Logical Memory (stories):
   Subtest WMS, STM -1.19 +1.52 -2.71 (-3.75 to -1.66) <0.001
   Subtest WMS, LTM -1.17 +1.70 -2.87 (-3.87 to -1.86) <0.001

Visual Reproduction
   Subtest WMS, STM -1.39 +0.04 -1.43 (-1.49 to -0.88) <0.001
   Subtest WMS, LTM -1.05 -0.11 -0.94 (-1.49 to -0.40) 0.003

*One-tailed test. P-values shown only if significant (P<0.05).

PASAT = Paced Auditory Serial Addition Task; WMS = Wechsler Memory Scale; STM = Short Term Memory; LTM = Long Term Memory

No significant differences were found in tests of speed of information processing (PASAT: 3.2 conditions, P=0.413) or attentional efficacy (Stroop Test: speed, P=0.07; number of mistakes, P=0.5). Tests that required fine-motor behavior proved to be invalid for this study because of the large intersubject variability.

Intragroup competitor comparison. Some traumatic exposure variables had a negative impact on cognitive function. Impact (weight in combination with the number of punches) reduced the number of categories (P=0.03) and number of elements reproduced in a visual memory Short Term Memory task (P=0.02). Knockouts and technical knockouts reduced the number of elements remembered in a verbal memory task. Lower ranked boxers (N and C levels) showed a slower response in the Stroop Test (P=0.01) compared with A- ranked boxers. One surprising result was that competitors who sustained a knockout had a better performance of the nondominant hand (P=0.04). There were no statistical differences in test performance between competitors who won, lost, or drew a match.


ATBI and studying cognitive impairment. Despite the requirement for protective headgear, there is a high risk for ATBI in amateur boxing. ATBI typically involves impairments in planning, attention, and memory, and our findings are in agreement with those of Critchley (27), who reported memory problems in boxers who had sustained an ATBI (knockout or technical knockout). Furthermore, Kelly and Rosenberg (7) documented attention and memory problems as acute symptoms in athletes who sustained concussions with or without loss of consciousness.

In this inception cohort study, several methodologic precautions were taken to minimize the problems associated with group comparisons. Cultural and educational differences were eliminated: Drug use, more than five alcoholic drinks per week, diagnosed learning disorders, and medical conditions that could affect cognitive functioning served as exclusion criteria. Controls were matched for age, boxing skill, and weight category, implicitly controlling for variables of previous boxing experience such as concussions sustained in previous fights and the number of fights. Moreover, the influence of possible confounding variables was limited by using the individual changes in cognitive functioning as outcome.

Physical activity variables were essentially equal for the two groups of boxers, and assessment and scoring procedures were controlled for expertise and bias. But differences in mental, physical, and emotional status could explain some of the results. Competitors fought a real match, evoking a "fight" response in physical, cognitive, and emotional systems that is difficult to replicate in the controls, who only punched a bag.

Both controls and competitors punched with both hands. The heaviest blows are delivered by the dominant hand, a fact that introduced greater intersubject variability in dominant-hand performance. As a consequence, the standard deviation of the Puncture Test (fine motor behavior) reflects this; it measures speed of fine motor behavior and accuracy. Force on the hand caused by delivering hard blows causes loss of fine motor speed and coordination (peripheral impairment). The standard deviation was high for the dominant hand but not for the nondominant hand.

ATBI in amateur boxing appears to be related to concussions, subconcussive blows, and trivial trauma associated with the impact of head punches. In this study, the amateur boxers sustained a median of eight head punches during the match, and 13% of the matches ended in concussions with or without loss of consciousness. Impact (weight in combination with the number of punches) of the blows delivered to the opponent in the match, knockouts, and technical knockouts were predictors for cognitive functioning after a match (impairment in memory and planning ability), while depression (losing the match) was not an important predictor. Moreover, cognitive symptoms of ATBI (planning, attention, and memory deficits) resemble those of CTBI. Thus, it is plausible that athletes who incur repeated ATBI do not fully recover from these injuries, and the continued insult may account for the insidious manifestations of CTBI in contact and collision sports.

In our study, competitors were less able to group items according to their semantic relatedness than were the controls. Such a finding demonstrates frontal lobe dysfunction; it is revealed by the decreased score in the categorization task. Levin et al (28) showed that patients with frontal lobe lesions had a lower ability to group related words as compared with both other-than-frontal lobe patients and normal controls; however, these results were obtained from nonathletes.

Recommendations. Our findings, that amateur boxing is associated with neuropsychological impairment, support the hypothesis that ATBI is a serious health problem for amateur boxers. Accordingly, prevention measures should be taken to maximize safety. The referee should stop the match at the first sign of distress, and the attending physician should have a mandate to stop fights immediately when a boxer shows ATBI symptoms. Concussion programs such as those advocated by the American Academy of Neurology (29) should be mandated by the Amateur International Boxing Association to reduce the cumulative effect of ATBI and to reduce the risk for diffuse cerebral swelling and second-impact syndrome (SIS).

SIS results from acute, usually fatal brain swelling believed to be due to recurrent head injuries when symptomatic from a previous ATBI (30,31). Although rare, sports-related SIS has a mortality approaching 100% (32,33). Although the designation of SIS is controversial (34), if the condition proves to represent a real clinicopathologic entity, rather than a subset of ATBI, it presents potential major consequences for the management of acute brain injuries in amateur and professional boxing. The Standardized Assessment of Concussion (SAC) as proposed by McCrea et al (35) and the test strategy to detect symptoms of concussion as proposed by Maddocks et al (36) are reasonable procedures for assessment of ATBI in amateur and professional boxing. We suggest, based on our study results, that visual memory tests should be added to the SAC. It should be noted that the SAC is not a surrogate for neuropsychological testing; however, the SAC has been adequately validated to serve that role when a neuropsychologist is not present and neuropsychological testing is not feasible, such as during or shortly after a sporting match. Visual memory tests are easy to administer and provide the physician with an impression of patient planning capacity.

Where Does Boxing Go From Here?

Participation in amateur boxing matches may impair cognitive functioning despite the use of protective headgear. This impairment appears to be attributable to the impact and number of concussions with or without loss of consciousness causing ATBI.

The duration and course of ATBI symptoms (eg, attention, memory, and planning) could not be documented in this study but are important issues for further research. Whether our findings can be extrapolated to other contact sports remains to be determined in future investigations.


  1. Jordan BD: Sports injuries, abstracted, in Proceedings of the Mild Brain Injury in Sports Summit. Dallas, National Athletic Trainers' Association, Inc, 1994, pp 43-46
  2. Ommaya A, Ommaya, Salazar A: A spectrum of mild brain injuries in sports, in Proceedings of the Mild Brain Injury in Sports Summit, Dallas, National Athletic Trainers' Association, Inc, 1994, pp 72-80
  3. Boden B, Kirkendall D, Garrett W: Concussion incidence in elite college soccer players. Am J Sports Med 192021;26(2):238-241
  4. Matser J, Kessels A, Jordan B, et al: Chronic traumatic brain injury in professional soccer players. Neurology 192021;51(3): 791-796
  5. Jordan BD: Boxing, in Jordan BD, Tsairis P, Warren RF (eds): Sports Neurology, ed 2, Lippincott-Raven, Philadelphia, 192021, pp 351-367
  6. McCunney R, Russo P: Brain injuries in boxers. Phys Sportsmed 120214;12(5):53-67
  7. Kelly JP, Rosenberg JH: Diagnosis and management of concussion in sports. Neurology 1997;48(3):575-580
  8. Evans R: The post concussion syndrome and the sequelae of mild head injury. Neurol Clin 1992;10(4):815-847
  9. McCrory P, Bladin P, Berkovic S: Retrospective study of convulsions in elite Australian rules and rugby league footballers: phenomenology, etiology and outcome. BMJ 1997;314(7075):171-174
  10. Drew R, Templar D, Schuyler H, et al: Neuropsychological deficits in active licensed professional boxers. J Clin Psychol 120216:42(3):520-525
  11. Matser J, de Bijl M, Luytelaar G. Is amateur boxing dangerous? De Psycholoog 1992;12:515-521
  12. Gronwall D, Wrightson P: Delayed recovery of intellectual function after minor head injury. Lancet 1974;2(7881):605-609
  13. Roberts A: Brain Damage in Boxers. London, Pitman Publishing, 1969
  14. Martland H: Punch drunk. JAMA 1928;91:1103-1107
  15. Johnson J: Organic psychosyndromes due to boxing. Brit J Psychiatr 1969;115:45-53
  16. Haglund Y, Eriksson E: Does amateur boxing lead to chronic brain damage? A review of some recent investigations. Am J Sports Med 1993;21(1):97-109
  17. Mendez M: The neuropsychiatric aspect of boxing. Int J Psychiatr Med 1995;25(3):249-262
  18. Gronwall D, Wrightson P: Cumulative effect of concussion. Lancet 1975;2(7943):995-997
  19. Jordan B, Matser J, Zazula T: Sparring and cognitive functioning in professional boxers. Phys Sportsmed 1996;24(5):87-2021
  20. Categorization task, in Kolb B, Whishaw I (eds): Fundamentals of Human Neuropsychology. New York City, WF Freeman, 1990, p 451
  21. Gronwall DM: Paced Auditory Serial Addition Task: a measure of recovery from concussion. Percept Motor Skills 1977;44(2):367-373
  22. Wechsler D: WAIS-R Manual, New York City, The Psychological Corporation, 120211
  23. Spreen O, Strauss E: A compendium of neuropsychological tests. New York City, Oxford University Press, 1991
  24. Golden CJ: Stroop Color and Word Test. Chicago, Stoelling, 1978
  25. Wechsler D: Wechsler Memory Scale Manual. San Antonio, TX, The Psychological Corporation, 1974
  26. van der Vlugt H: Lateralisaties van hersenfuncties: een neuropsychologisch onderzoek naar de relatie tussen handvoorkeur en de relatie van de taalfunctie [Neuropsychological research of the puncture test for measurements of fine motor behavior. Dissertation]. Lisse, The Netherlands, Swets and Zeitlinger, 1979
  27. Critchley M: Medical aspects of boxing, particularly from a neurological standpoint. BMJ 1957;1:357-362
  28. Levin H, Goldstein F, Williams D, et al: The contribution of frontal lesions to neurobehavioral outcome of closed head injury, in Levin H, Eisenberg H, Benton A (eds): Frontal Lobe Function and Dysfunction. New York City, Oxford University Press, pp 318-338
  29. Quality Standards Subcommittee: Practice parameter: the management of concussion in sports. Neurology 1997;48(3):581-585
  30. Kelly JP, Nichols JS, Filley CM, et al: Concussion in sports: guidelines for the prevention of catastrophic outcome. JAMA 1991;266(20):2867-2869
  31. Cantu RC, Voy R: Second impact syndrome: a risk in any contact sport. Phys Sportsmed 1995;23(6):27-34
  32. McQuillen JB, McQuillen EN, Morrow P: Trauma, sport and malignant cerebral edema. Am J Forensic Med Pathol 120218;9(1):12-15
  33. Saunders RL, Harbaugh RE: The second impact in catastrophic contact-sports head trauma. JAMA 120214;252(4):538-539
  34. McCrory P, Berkovic S: Second impact syndrome. Neurology 192021;50(Mar):677-683
  35. McCrea M, Kelly MD, Kluge J, et al: Standardized assessment of concussion (SAC) in football players. Neurology 1997;48(3):586-588
  36. Maddocks DL, Dicker GD, Saling MM: The assessment of orientation following concussion in athletes. Clin J Sports Med 1995;5(1):32-35

Dr Matser is an assistant professor in the department of anatomy, faculty of medicine and health sciences, Erasmus University, in Rotterdam, The Netherlands. Dr Kessels is an epidemiologist in the research unit patient care, University Hospital of Maastricht, The Netherlands. Dr Lezak is a neuropsychologist in the department of neurology, Oregon Health Sciences University, Portland, Oregon. Dr Troost is a neurologist in the department of neurology, University Hospital of Maastricht, The Netherlands. Dr Jordan is a neurologist in the traumatic brain injury program, Burke Hospital, White Plains, New York. Address correspondence to Erik J.T. Matser, PhD, Dept of Sports Neurology, St Anna Ziekenhuis, Box 90, 5660 AB, Geldrop, The Netherlands; e-mail to [email protected].