Abstract
Background: Few studies have reported acute postbout cognitive function in amateur boxers, and none have documented the effects of repeated boxing bouts within a short time frame.
Objective: To determine whether participation in a 7-day amateur boxing tournament is associated with acute deterioration in cognitive test performance.
Methods: A prospective study was done of 82 collegiate amateur boxers participating in a 7-day single elimination tournament and a group of 30 matched nonboxing control participants. No participants had a history of recent concussion or past history of brain injury. For boxers, cognitive assessment using a computerized test battery was performed before the tournament and within 2 hours of completing each bout. Tests of simple and choice reaction time, working memory, and learning were administered. Analysis of variance was conducted to compare the serial performance of control participants with that of boxers participating in one, two, and three bouts.
Results: The 82 boxers fought 159 times. Cognitive testing was performed after 142 of these bouts. On simple reaction time, choice reaction time, and working memory tasks, the serial performance of boxers participating in three bouts (n = 22) was equivalent to that of boxers participating in two bouts (n = 22) and one bout (n = 32) and to nonboxing control participants (n = 30). An improvement in performance was observed on the learning task in boxers participating in three bouts. Boxers whose bout was stopped by the referee (n = 7) displayed significant slowing in simple and choice reaction time.
Conclusions: With the exception of boxers whose contest is stopped by the referee, amateur boxers participating in multiple bouts during a 7-day tournament display no evidence of cognitive dysfunction in the immediate postbout period.
It is well known in professional boxing that a variety of chronic neurocognitive sequelae may accompany participation in this sport. These include the clinical effects of chronic traumatic encephalopathy (or the “punch drunk” syndrome),1–3⇓⇓ neuropathologic injury,4 and cognitive impairment.5 It has been speculated that many of these neurologic and cognitive phenomena may reflect a genetic risk rather than simply exposure to head impact.6
In amateur boxing, the exposure to repeated head impact is less than that seen in professional boxing, principally because of the shorter duration of the bouts and the use of protective headgear.7 Nevertheless, chronic traumatic encephalopathy has been documented in this population, albeit at a reduced frequency when compared with professional boxing.4,8⇓ There have been a number of studies demonstrating the lack of acute cognitive effects after a single amateur boxing bout.9–19⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓ However, two studies of boxers diagnosed with concussion suggest that participation in amateur boxing bouts may decrease neurocognitive function.20,21⇓
One particular concern in boxing is the role of repeated subconcussive blows in the genesis of chronic neurologic injury. One study of young boxers demonstrated an association between declining neuropsychological test performance and increased exposure to sparring, suggesting that repeated subconcussive blows to the head may be as damaging as blows causing acute concussive symptoms.22 There is conflicting evidence from other sports, such as soccer, that increasing exposure to situations in which head impact may occur may similarly be associated with decrease in cognitive function.23–25⇓⇓
Amateur boxing is typically conducted in a tournament situation, in which boxers may participate in numerous bouts during a period of days or weeks. Despite this, there have been no published studies investigating the cognitive or neurologic consequences of participating in repeated boxing bouts during a short period. We sought to determine whether participation in a single bout and in multiple bouts is associated with acute postbout cognitive dysfunction. The primary comparison was between individuals participating in single and multiple bouts, using number of bouts as a de facto measure of exposure to head impact. Further, this study sought to investigate immediate postbout cognitive function in boxers who were at high risk for cognitive impairment caused by nonconcussive events occurring during the boxing bout (e.g., epistaxis, standing eight count, and referee-stopped contest [RSC]).
Methods.
Participants.
The Notre Dame Bengal Bouts is a collegiate boxing tournament held annually on the campus of the University of Notre Dame in Indiana. In 2002, 135 members of the Notre Dame Boxing Club participated in the tournament. Of these, 85 boxers consented to participate in this study. All boxers, including the study participants, were required to undergo a physical examination and answer a standardized medical questionnaire. Any boxer reporting a history of severe head injury, bleeding disorder, diabetes, or recent concussion was precluded from participating in the tournament. A group of 30 nonboxing, age-matched university students were also enrolled to provide control data for statistical analyses. These participants were assessed at the University of Melbourne in Australia. All participants were male and were tertiary educated. The respective university human ethics committees approved the study design. Informed consent was obtained from all participants.
Materials.
A computerized neuropsychological test battery was used in this study (CogSport, CogState Ltd., Melbourne, Australia). A full description of this battery can be found elsewhere.26 CogSport is a series of computerized card tasks that requires 15 to 18 minutes to complete. The test battery includes seven distinct tasks, four of which were selected for analysis in this study: simple reaction time, choice reaction time, working memory, and learning. These cognitive domains were chosen for inclusion based on previous work demonstrating susceptibility to mild head injury and concussion.27–31⇓⇓⇓⇓
This computerized test battery has demonstrated sensitivity to mild cognitive changes caused by concussion,32 fatigue,33 alcohol,33 early neurodegenerative disease,34 coronary surgery,35 and malingering/simulation.36 The effects of practice37 and the reliability of the test26 have been documented, as has its correlation with conventional paper-and-pencil tests.26 Other metric properties of the test have also been reported.38
Procedure.
Participants (boxers and control subjects) were required to practice the computerized test battery before submitting a baseline test.39 For boxers, all baseline tests were conducted in a computer laboratory 2 weeks before the tournament. Control participants also completed baseline tests in a university computer laboratory. All baseline test results were compared with previously published age-, education-, and gender-appropriate normative data to determine whether each participant was performing at an appropriate level.26 Any participant whose baseline performance was “below the normal range” (normative group mean − 1.96 SD) on any task was required to complete a second baseline test. Any participant whose performance on the second baseline test remained below the normal range was excluded from the study. Control participants returned to the university computer laboratory exactly 2 weeks after the baseline test to perform follow-up testing.
The single elimination tournament was spread over 7 days with boxing on days 1, 3, 5, and 7 and consisting of preliminaries, quarterfinals, semifinals, and finals, respectively. Boxers were stratified into 10 weight divisions. There were an uneven number of participants in each weight division; therefore, some boxers were not required to box preliminary or quarterfinal bouts. Bouts were fought according to USA Amateur Boxing Federation (USA/ABF) rules with the winners of each bout advancing to the next round. For the preliminary, quarterfinal, and semifinal bouts, three rounds of 1.5-minute duration were fought with 50-second breaks between rounds. Final bouts consisted of three 2-minute rounds with 1-minute breaks between rounds. All boxers wore hand wraps under their gloves. Boxers weighing <170 pounds (77.3 kg) wore 14-ounce gloves. Boxers weighing >170 pounds wore 16-ounce gloves. All boxers wore headgear. The ringside physician or the referee could stop the bout any time a boxer appeared impaired or unable to defend himself. Ringside physicians kept a log of all “high-risk” bouts, in which there were significant head blows (as defined by the boxing judges), concussions, RSC, standing eight counts, or epistaxis.
All boxers were screened for evidence of concussion by certified athletic trainers immediately after their bouts. Screening consisted of observation of the boxer’s demeanor, evaluation of facial trauma, orientation, memory loss, and symptoms suggestive of head injury. Any boxer suspected of sustaining a possible concussion and all high-risk boxers (as described previously) were referred for physician evaluation. Concussion grading scales were not used because of their lack of scientific validation.40 For the purposes of this study, concussion was defined as head trauma resulting in alteration in mental state or the onset of clinical symptoms, or both, and was diagnosed based on the physician’s clinical evaluation of the athlete. After physician evaluation, boxers were then orally rehydrated and allowed to rest for at least 30 minutes before the cognitive assessment. An experienced physician (J.M.) supervised cognitive testing. All cognitive tests were administered within 2 hours of bout completion. Tests were administered on laptop computers in a controlled, quiet environment. At baseline and follow-up assessments, all participants wore headphones to eliminate distracting noise.
Data analysis.
The number of boxers completing baseline and postbout cognitive assessments, number of bouts, number of assessments, and number of high-risk bouts were determined. These data are described in text, below. Because the major hypothesis concerned the overall acute effects of participation in an amateur boxing tournament, cognitive test results obtained after high-risk bouts were included in the group analyses. However, these results were also analyzed separately according to the type of high-risk event (see below).
It is common practice with measures of reaction time to “trim” the data of outlying responses before analysis.41 However, because previous studies have observed increased response variability after head injury, such outliers were included in the current analysis.27,28⇓ For each participant, anticipatory responses (defined as responses faster than 100 ms) were counted as errors and excluded from further analysis. The mean reaction time on each task was used to express the speed of performance. Inspection of the distributions of reaction times indicated a positive skew in all distributions. This is a common feature of reaction time distributions.41 Therefore, data for each participant were logarithmic base 10 (log10) transformed before statistical analysis to ensure that data met the assumptions of normality and heterogeneity of variance. Untransformed reaction time data are presented in the tables; transformed data are available on the Neurology Web site (tables E-1 and E-2; go to www.neurology.org). Inspection of accuracy data for all four tasks indicated that ceiling effects were evident for simple reaction time, choice reaction time, and working memory tasks. Consistent with previous analysis using this computerized test battery, we therefore chose to present accuracy data for the continuous learning task only.33,36⇓ All accuracy data were arcsine transformed before statistical analysis; however, untransformed data are presented in the tables.
Changes between baseline and post-last-bout cognitive performance of boxers participating in three bouts was compared with those of boxers participating in one or two bouts and with control group data using a group (4; control, 1 bout, 2 bout, and 3 bout) by assessment (1; baseline, post-final bout) repeated-measures analysis of variance (ANOVA) for each task. When the group by assessment interaction was significant, post-hoc Dunnett test was used to directly compare the three boxing groups with the control group.
A second analysis sought to determine whether significant acute cognitive changes occurred in boxers who were considered at high risk for cognitive impairment. For all measures, simple change scores were calculated to determine the magnitude of change between baseline and postbout performance in those 28 instances in which a high-risk event was noted.42 One-way ANOVA with post-hoc Dunnett t-test was used to compare the change scores of individuals within each high-risk group with scores from control participants. For the accuracy measure, a group (5; control, RSC, epistaxis, hard blow, and standing eight count) by time (2; baseline, postbout) repeated-measures ANOVA was conducted. When the group by assessment interaction was significant, post-hoc Dunnett tests were conducted to directly compare the high-risk groups with the control group.
Results.
All 85 boxers passed their physical examination and medical questionnaire. Of this group, 82 achieved baseline CogSport testing judged to be adequate for inclusion in the study (i.e., within the normal range). These 82 participants fought 159 bouts, and cognitive testing was performed after 142 of these bouts (90.4% compliance). The 17 “missed opportunities” were distributed among 14 boxers and were a result of the boxer’s unwillingness to perform the test. None of the missed opportunities resulted from boxing injury.
Frequency of testing for each boxer depended on the boxer’s progression in the tournament. In total, 32 boxers had one bout and performed two tests (including baseline); 22 boxers had two bouts and performed three tests; and 22 boxers had three bouts and performed four tests. The ages of boxing and control groups are described in table 1.
Table 1 Participant ages
Group mean and SD performance of control subjects and boxers who participated in one, two, and three bouts is presented in table 2. Transformed data are presented in table E-1 (go to www.neurology.org). Repeated-measures ANOVA demonstrated that there were no group by task interactions on simple reaction time (p = 0.62), choice reaction time (p = 0.25), working memory (p = 0.12), and continuous learning accuracy (p = 0.33) variables. However, an interaction was observed for the speed of performance on the continuous learning task (F(3,103) = 2.76; p = 0.03). Post-hoc Dunnett t-tests revealed that the learning task performance of boxers participating in three bouts improved compared with control participants (p < 0.01). This is reflected in a moderate Cohen d value of −0.80 displayed in table 2.
Table 2 Baseline and follow-up performance of boxers and control participants
Additional t-tests conducted on learning task data for boxers with three bouts revealed no changes in performance from baseline to bout 1 (p = 0.08), bout 1 to bout 2 (p = 0.15), bout 2 to bout 3 (p = 0.84), and bout 1 to bout 3 (p = 0.08). However, performance changes were observed between baseline and bout 2 (t(21) = 5.25; p < 0.001) and between baseline and bout 3 (t(21) = 4.74; p < 0.001).
Twenty-eight high-risk bouts were recorded. Eleven of these involved standing eight counts; seven involved RSC; five involved epistaxis; and four involved the boxer receiving a remarkably hard blow as rated by ringside judges (table 3). One boxer diagnosed with concussion was excluded from analysis because he was unavailable for assessment within the immediate postbout period. One-way ANOVA was significant for the simple reaction time (F(4,53) = 3.12; p = 0.02) and choice reaction time tasks (F(4,53) = 50.11; p < 0.01). Post-hoc t-test revealed that boxers with RSC had a significant decline in performance compared with control boxers on the simple reaction time (p < 0.01) and choice reaction time tasks (p < 0.01). In addition, boxers with epistaxis displayed a slowing of choice reaction time compared with control participants (p < 0.01). No other significant interactions were observed on ANOVA. Transformed data are presented in table E-2 (go to www.neurology.org).
Table 3 Baseline and postbout performance of “high-risk” athletes
Discussion.
Participation in multiple amateur boxing bouts during a medically supervised and strictly controlled week-long tournament is not associated with acute cognitive impairment in collegiate athletes. When compared with their own performance at a baseline assessment and with the serial performance of a matched control group, the postbout performance of 22 amateur boxers participating in three boxing bouts during a 7-day period did not decline on computerized neuropsychological tests of simple and choice reaction time, working memory, and learning accuracy. Further, this group displayed an improvement in performance on a learning speed task, suggestive of practice effects caused by the repeated administration occurring over a relatively short period.34,37⇓ Amateur boxers participating in two bouts and in a single bout showed no changes in cognitive test performance.
Analysis of boxers considered at high risk for cognitive impairment revealed no performance impairments in boxers receiving standing eight counts and hard blows. Seven boxers whose bouts were stopped by the referee but who were not clinically diagnosed with concussion demonstrated a significant slowing of performance on a simple reaction time and choice reaction time tasks. Further, boxers with epistaxis displayed a significant slowing of performance on a choice reaction time task. These data are difficult to interpret; however, in the context of previous studies suggesting that simple and choice reaction time tasks are more sensitive to brain injury than more complex neurocognitive tasks, these results cannot be dismissed and deserve further investigation.21,27,28⇓⇓ Only one concussion was observed in 159 matches. Consistent with clinical management guidelines, this boxer was retested 1 week after his bout while asymptomatic, and no cognitive impairments were observed.40 This is consistent with the previous literature and clinical experience.32
We believe that this is the first study to prospectively examine the serial cognitive performance of nonconcussed amateur boxers participating in a tournament situation. This is significant because competitive amateur boxing is typically conducted in tournaments such as that described here. Other recent studies have examined neuropsychological performance of amateur boxers clinically diagnosed with concussion. For example, one study evaluated 483 US Military Academy cadets participating in physical education boxing with Automated Neuropsychological Assessment Metrics (ANAM).21 Twelve cadets sustained head injury severe enough to be diagnosed as concussion. All had delayed simple reaction times at 1 hour and 4 days postinjury compared with baseline. All were asymptomatic 24 hours after injury. However, no analysis was performed on boxers without a concussion diagnosis. Another study of 38 Dutch amateur boxers using postmatch paper-and-pencil neuropsychological testing demonstrated acute but mild cognitive impairment on measures of planning, attention, and memory.20 The bouts in this study were described as having frequent concussions, knockouts, and memory impairment. It is possible that the safety precautions, supervision, and early termination of bouts before more severe head trauma in our study accounted for the differences in cognitive outcome. This is supported by the low incidence of concussive injury observed in the present study relative to these previous studies. It is also important to note that other published studies have demonstrated an absence of cognitive dysfunction after a single boxing bout.9–19⇓⇓⇓⇓⇓⇓⇓⇓⇓⇓ The current study further suggests that participating in multiple bouts in a tournament situation does not result in acute cognitive dysfunction unless the contest is stopped by the referee.
The present results do not address issues of chronic cognitive impairments in amateur boxers nor may they be generalized to professional and other forms of boxing and martial arts or to amateur boxing tournaments in which different rules, regulations, and medical procedures are in place. We await well-designed, prospective long-term studies of cognitive and neurologic function in amateur and professional boxers with great interest. Evidence of a practice effect associated with repeated administration was observed on the learning task in boxers participating in three bouts. This is consistent with previous literature suggesting that increased exposure to testing and task complexity affect the presence and magnitude of practice effects and represents a limitation of the present study.37,43⇓ One further caveat of this study is that an indirect measurement of exposure to head injury was used (number of bouts). Biomechanical analysis (e.g., via video analysis) of the cumulative force of blows during such bouts is required to draw direct association between the extent of subconcussive head blows and postbout cognitive function. Finally, although performance was equivalent between control participants and boxers, these two groups were drawn from distinct geographic areas, and boxers with multiple bouts were also assessed on more occasions than control participants. This latter point may have led to the practice effect observed in boxers with three bouts on the continuous learning task.
This study has implications for amateur boxing worldwide. It appears that as long as appropriate medical and safety procedures are in place, boxers may participate in amateur boxing tournaments with little risk of acute cognitive impairments. In broader terms, stringent medical treatment of boxers before, during, and after an amateur boxing tournament may reduce the risk of acute neurologic injury. Boxers whose contests are stopped by the referee and those with epistaxis should be considered to have acute cognitive impairments until proven otherwise.
Footnotes
Additional material related to this article can be found on the Neurology Web site. Go to www.neurology.org and scroll down the Table of Contents for the May 11 issue to find the title link for this article.
See also page 1462
For the duration of the study, Dr. Alex Collie and Michael McStephen were employees of CogState Ltd., the manufacturer of the cognitive testing software used in this study.
- Received July 1, 2003.
- Accepted in final form February 23, 2004.
References
Letters: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Hemiplegic Migraine Associated With PRRT2 Variations A Clinical and Genetic Study
Dr. Robert Shapiro and Dr. Amynah Pradhan