Article Text

Download PDFPDF

Low risk of concussions in top-level karate competition
  1. Rafael Arriaza1,
  2. Dusana Cierna2,
  3. Patricia Regueiro3,
  4. David Inman4,
  5. Franco Roman5,
  6. Benjamin Abarca6,
  7. Mercé Barrientos3,
  8. Miguel A Saavedra3
  1. 1Instituto Médico Arriaza y Asociados, Cátedra HM de Traumatologia del Deporte, Universidade da Coruña, A Coruña, Spain
  2. 2Physical Education and Sports School, Comenius University, Bratislava, Slovak Republic
  3. 3Physical Education and Sports School, Universidade da Coruña, A Coruña, Spain
  4. 4Oldfield Osteopathic Clinic, Bath, UK
  5. 5Commission Médicale Nationale de la Fédération Française de Karaté et Disciplines Associées, Paris, France
  6. 6Comisión Médica Nacional, Federación Chilena de Karate, Puerto Montt, Chile
  1. Correspondence to Dr Rafael Arriaza, Instituto Médico Arriaza y Asociados, Cátedra HM de Traumatologia del Deporte, Universidade da Coruña, Calle Enrique Mariñas, 32, A Coruña 15008, Spain; rafael{at}


Background Although it is well known that injuries occur in combat sports, the true incidence of concussions is not clearly defined in the literature for karate competition.

Aim To determine the incidence of concussions in top-level (World Karate Federation World Championships) karate competition.

Methods Injuries that took place in 4 consecutive World Karate Championships (from 2008 to 2014) were prospectively registered. A total of 4625 fights (2916 in the male category and 1709 in the female category) were scrutinised, and concussions were identified and analysed separately for frequency (rate per fight) and injury risk.

Results A total of 4 concussions were diagnosed by the attending physicians after carrying out athlete examinations. Globally, there was 1 concussion in every 1156 fights, or 0.43/1000 athlete-exposures (AE). In male athletes, the rate of concussion was 1/5832 min of fighting, and in female athletes, it was 1/6836 min. OR for concussion in women is 0.57 (95% CI 0.06 to 5.47; z=0.489; p=0.6249) and risk ratio for concussions in men is RR 1.478 (95% CI 0.271 to 8.072), p=0.528, representing a higher risk of definite concussions in men than in women, but not statistically significant. There is not a significantly higher risk of concussions in team competition (no weight limit) when compared with individual competition (held with strict weight limits for each category).

Conclusion The risk of concussions in top-level karate competition is low, with a tendency for an increased risk for men and for competition without weight limits, but not statistically significant with respect to women or individual competition.

  • Contact sports
  • Injuries
  • Concussion

Statistics from


Sports that involve contact and/or collisions, such as boxing, American football, ice hockey, soccer, rugby and martial arts, are associated with an increased risk of traumatic brain injuries.1 The World Karate Federation (WKF) and its competition rules are recognised by the International Olympic Committee. This shall be defined as ‘Olympic karate’, to avoid confusion with other minor competition styles and rules, in which the rate of concussions may be different. The rules of some full contact sports, such as taekwondo, kickboxing and mixed martial arts, allow for face and head contact aiming to knock down the opponent This is not the case in Olympic karate, where only minimal contact to the head is allowed. However, accidental head collisions and excessive contact may occur causing injury to participants. These collisions may lead to injuries, including concussions.2

The general epidemiology of concussions in boxing and other combat sports has been studied but, owing to the wide variations in karate styles and competition levels, the incidence of concussions has been only grossly defined, within a range of 0.9–5.5% of the total number of injuries.3 Nevertheless, the incidence of concussions in Olympic top-level karate competition is unknown and, as has been stated by other authors, in order to develop evidence-based, sport-specific management guidelines, detailed information is required about the magnitude, nature and causes of concussion within individual sports.4

There are variations in the definition of concussion, and various guidelines have been proposed.5 Neurological imaging and laboratory tests have not revealed corresponding abnormalities, which makes diagnosis somewhat elusive, and are based mainly on clinical findings.6

Owing to the lack of objectivity in the diagnosis of concussions, it is important to have data from different sports and also from different levels of competition within each sport. This should provide a basis for comparison and guide the physicians involved in competitive combat sports, thus progressing our knowledge of the epidemiology of concussions in sports. Combat sports share some common characteristics, such as division of competition by sex and weight. Karate is unique among combat sports because there is a free-weight team competition category, in which teams of competitors (5 for men and 3 for women) fight in a random order, potentially allowing athletes with different body weights to compete against each other. Also, it is a sport very popular among young women, and it is important to investigate if there is a higher risk of concussions in female than in male categories, due to the paucity of data regarding female athletes engaged in combat sports,7 and also if the categories without weight divisions (ie, ‘teams’) have a higher concussion risk than the categories with strict weight divisions (ie, ‘individuals’).

The purpose of this study was to define the rate of concussion in top-level karate competition in the WKF from 2008 to 2014, a period of time where competition rules did not change, to obtain a higher homogeneity in data.

The hypothesis was that male and female competition should have a different risk of concussions and that ‘team’ competition (confronting athletes with different body weights) would have a higher risk of concussion than ‘individual’ competition (where athletes are grouped under strict weight limits).

Materials and methods

Championships and fights included in the study

The study followed a prospective cohort design, and included data collected from October 2008 to November 2014, during four consecutive WKF World Championships. All the athletes who entered the four World Karate Championships and all the bouts associated with them were retrieved from the official drawing data sheets of the championships, which allowed for a precise competitors and match count.

Injury recording and concussion assessment

During the four World Karate Championships studied, injuries were recorded following the same protocols used in previous investigations.2 ,8 According to karate competition rules, whenever there was a suspected injury on the tatami (the competition area) during the studied World Championships, the doctor in charge was called by the referee to assess the competitor and, if necessary, provide treatment. For each tatami (individual competition area), there was one doctor plus one assistant responsible for medical support. In addition to caring for injured athletes, the medical teams at each championship were also responsible for recording all injuries that took place during the competition. Every injury that was seen by the tournament doctors was recorded, no matter how minor. Data were collected by the attending doctors as soon as the injuries took place, using check off forms that described the competitor's country, sex, category and weight, the injured area, diagnosis, mechanism of injury, severity and treatment. No names or other personal information was registered, to guarantee anonymity.

Prior to the initiation of each championship, the senior author, RA (as Chairman of the Medical Commission of the WKF), instructed the local doctors and physician assistants on the injury recording procedure. He also reviewed the recorded data in situ during the championships, three times per day, to clarify any possible doubts and to use a uniform injury classification. From the data, concussions were identified and analysed separately for frequency (rate per fight) and injury risk (as concussions per 1000 athlete exposures and as concussions per 1000 min of fighting).

For any athlete suspected of having suffered a concussion, it was mandatory that he/she was examined by two members of the WKF Medical Commission, to determine if he/she was allowed to continue in the competition, or was sidelined for the rest of the tournament as stated in the WKF Competition Rules Among the members of the WKF Medical Commission who supervise the World Championships, there is one Neurologist and one Neurosurgeon, which allows for immediate in situ neurological evaluation of any karateka suspected of having suffered a concussion.

Concussions were defined when, after an impact to the head, an athlete showed loss of consciousness (LOC), loss of balance, amnesia, dizziness or disorientation: a modification of Maddock's questions adapted to a combat sport in which only one round is fought per bout, and balance tests were used (see online supplementary appendix 1), following the SCAT-1 and SCAT-2 guidelines.5 ,10 When an athlete suffered an impact that required medical assessment and where concussion was suspected, but examination did not disclose definite signs to confirm the diagnosis, it was registered under ‘suspected concussion’. The athlete was then closely scrutinised by the medial team and the referee panel in the following bouts, in order to protect him/her if a new impact to the head should occur, or if a previous concussion was undiagnosed or manifested subsequently. Any concussion detected (no matter whether accompanied by LOC or not) implied the withdrawal of the competitor from competition. Simultaneously, as stated in the WKF competition rules, injury to one competitor caused by another may lead to the disqualification of the offending opponent. This requires the precise judgement by referees and the diagnostic confirmation of the attending physician.

By entering the championships, athletes (who were all over 18 years of age) agreed to allow for injury registration and treatment.

Concussion risk analysis

Each bout lasted 3 real-time minutes in male categories, and 2 real-time minutes in female categories, and so it was decided to analyse the risk of injuries per fight (as a clear measure of athlete-exposures (AE), as is performed nowadays in most sport epidemiology studies), and also per 1000 min of active fighting, to correct for the different exposure times of male and female athletes. Results were analysed separately for categories in which the competitors were divided according to strict weight categories (individuals) and categories in which athletes of different weights could be fighting each other (team competition), as there is the possibility that body mass difference may represent a risk factor that should be considered by competition rules in the future.

Statistical analysis was conducted to identify risk factors for the presence or absence of injury, including sex, weight category and championship. All data were analysed in statistical software IBM SPSS (V.21 for Windows). Cohorts were compared by computing the rate ratio (RR) of two injury incidence risks (IIRs). Ninety-five per cent CIs were computed for all RRs and relative risks using standard formulae for the Poisson rates and binomial proportions. The CIs for relative risks were used to determine whether two rates or proportions differed significantly from one another. Phi and Cramer's correlation were used to assess the significance of each predictor. ORs were computed for each risk factor as a measure of effect and 95% CIs were constructed. A priori α was set at 0.05.

The Ethical Committee on Research of the University of A Coruña approved the study (UDC EC-12/2016).


Data from a total of 4625 fights (2916 in the male category and 1709 in the female category) in the World Karate Championships were prospectively registered and retrospectively analysed. They represent all the fights that have taken place in the four World Championships held from 2008 to 2014, and accumulate 12 166 min of fight (table 1).

Table 1

Championships included in the study (city and year) and number of individual and team fights (bouts) and minutes of fighting (note that male fights last 3 min and female fights last 2 min) and diagnosed concussions (DC)

A total of 15 suspected concussions were identified during the 4 World Championships studied. Of these, four cases were diagnosed as definite concussions by the attending physicians, using the criteria previously stated in online supplementary appendix 1, following the SCAT-1 and SCAT-2 guidelines.5 ,10 The remaining 11 cases were deemed to be non-concussive injuries. Table 1 shows the distribution of the concussions diagnosed by the medical teams during the different championships analysed.

Results of male and female competition (in individual and team events) are shown in table 2. Nine of the 15 suspected concussions took place in individual competition, and 6 occurred in team competition; OR 1.095 (95% CI 0.389 to 3.078), p=0.864; RR 0.964 (95% CI 0.637 to 1.457), p=0.542, representing a tendency to increased risk of suffering a head impact not causing a concussion in team (no weight-limit) competition with regard to individual competition, but not statistically significant. Globally, the rate of suspected concussion in male athletes was 1/1750 min of fighting, and in female athletes, it was 1/1367 min. OR was 0.853 (95% CI 0.291 to 2.497), p=0.77; RR was 1.109 (95% CI 0.542 to 2.269), p=0.50, representing an increased tendency to suffer head impacts not causing concussions in women than in men, but not statistically significant.

Table 2

Risk of suspected and definite concussion per fight and per 1000 AE in men and women

Three of the definite concussions took place in team events (two in male fights and one in a female fight), and only one in individual events, for an OR of 0.365 (95% CI 0.033 to 4.023), p=0.410 and an RR of 1.735 (95% CI 0.350 to 8.596), p=0.384, representing an increased tendency of concussions in team (non-weight-limits competition) than in individual fights. In male individual competition, there was one concussion, and in male team competition, there were two concussions, for an OR of 0.398 (95% CI 0.036 to 4.387), p=0.452 and an RR of 1.671 (95% CI 0.337 to 8.281), p=0.415, representing a tendency to a higher risk of concussions in male team competition, but it was not statistically significant. There was an increased tendency of suffering a head impact not causing a concussion in female individual competition than in female team competition, but it was not statistically significant (for an OR of 2.513 (95% CI 0.81 to 22.511), p=0.393; and a RR of 0.768 (95% CI 0.495 to 1.191), p=0.362). In female individual competition, there were more concussions than in team competition, but it is not possible to calculate risks due to the low numbers: female team 0.76/1000 AEs (1/2636); female individual 0/1000 AEs (0/2100). Neither in men nor in women is there a significantly higher risk of injury in team competition (no weight-limit) than in individual competition. Also, the risk of suffering a concussion was not significantly different for men or women competitors.


The main finding of this study is that the risk of a possible concussion in Olympic karate top-level competition is low (1.62 suspected concussions per 1000 AE), and even lower for definite concussions (0.43 per 1000 AE). There are small differences in the risk for male and female athletes, team (non-weight restrictions) and individual (strict weight limits) competition, with a small tendency to suffer more concussions in male athletes and team events, although not statistically significant. Thus, our initial hypothesis that the risk of concussions would be higher in combat categories where athletes are not separated by weight was not proved. A possible explanation is that Olympic karate is not a full contact combat sport and offers a greater protection to the athletes than that which may have been predicted.

When compared with other combat sports, we must bear in mind that Olympic karate is not a full contact sport, and that direct risk comparison is somewhat misleading. In that sense, the systematic review of Lystad about the risk of concussions in Olympic Taekwondo showed a risk of 4.9 concussions per 1000 AE, which represents a risk around 10 times higher than the one found in our study for Olympic karate (a rate of 0.43 concussions per 1000 AE).11 The risk in that study was even higher in adolescent and young athletes. Although it is logical that the injury risk (and especially that of a concussion) is higher in full contact combat sports than in semicontact sports,12 we believe that this information is very important when advising athletes and families on the risks of entering into one sport or another. Also, this information should help the federation governing bodies to implement or modify competition rules and protective equipment, in order to minimise the risk of potentially severe injuries, such as concussions.

One of the main problems related to sport concussions is the difficulty of on-field diagnosis and management.13 A considered strength of the results obtained in this article is that an athlete was examined by a doctor at the side of the tatami, observing the bout; also, the diagnosis was confirmed by further examination from two more doctors, highly experienced in combat sport medicine and neurology (belonging to the WKF Medical Commission). WKF competition rules make it compulsory that at least one doctor plus one assistant attend each competition area or tatami during international championships. This and the use of the same injury collecting systems has been in place for over 20 years and has allowed a precise count of the injuries that take place during top-level WKF karate competition in a prospective manner, minimising the loss of information. Logically, one of the limitations of this study is that there might have been undetected concussions, as a result of athletes trying to hide symptoms, in order to avoid being medically retired from competition. However, this is something that can be argued in every study analysing the risk of concussion in sports,14 especially considering that although the term concussion is widely used, there is not one universally accepted definition of concussion. Another limitation is the fact that previous concussion history of the athletes was not available, and thus the possible presence of a risk factor cannot be quantified. Also, it is a fact that athletes know that if they are diagnosed with a concussion, they will be retired from the competition. This may discourage some of them from fully disclosing diagnostic clues to the doctors that might have led to a diagnosis of concussion. No athlete suffered a second concussion (competition category and nationality were different among the injured athletes) and in the case of suspected concussions, a mark was made on the identity card of the competitor. This mark was clearly visible to the referees and the medical personnel in charge of the competition areas, and informed them that the athlete should be medically retired if they sustained another impact to the head. No such cases were identified in this article.

In karate, concussions are the result of forces transmitted to the head, through accidental contact with the opponent's padded fist, foot or, more rarely (due to competition rules), elbow or leg and, even more rarely, the result of the impact of the head against the (padded) floor. Concussions often result in rapid, but short-lived, impairment of neurological function. These clinical symptoms are often the result of functional disturbances and not structural injury. Thus, traditional imaging modalities (eg, MRI or CT scans) often result in negative findings when diagnosing an athlete with concussion and clinical examination is paramount to allow proper diagnosis.15 LOC is not a requirement for diagnosis of concussion. Most concussions in adults tend to resolve spontaneously within 7–10 days,16 although the recovery period can be longer in children and young adolescents. It is possible that in our study, some of the injuries defined as suspected concussions were in fact concussions that had passed unnoticed, because the symptoms had completely resolved when doctors arrived at the tatami (although the average distance from the doctors' table to the tatami is only 10 m). At all times, the attending doctors adhered to strict diagnostic criteria and were not influenced by the referees or the athletes’ coaches. As such, the integrity of the WKF medical team was maintained in providing safety for the athletes and to reduce the risk of suffering further impacts that could give rise to severe brain damage.17

The results from a study of concussion by the National Football League demonstrated no cases of second impact syndrome or catastrophic head injury in players returning to play in the same game after resolution of symptoms,18 but studies of professional boxers have shown that repeated brain injury can lead to chronic encephalopathy, termed dementia pugilistica.19 Evidence from other sports that involve repetitive head impact, such as soccer, have demonstrated that players who have sustained multiple minor concussions performed worse on neuropsychological tests compared with a control group.20 ,21 Concussion grading and classification has evolved over time, and actual recommendations have moved away from the importance given in the past to the LOC,22 and so we decided to include in our study the head impacts that led to medical examination of the athletes, even when a definite diagnosis of concussion was not reached. For this reason, during the championships analysed, the identity cards of the competitors with suspected concussions were marked to have a closer follow-up, even if they did not show the full set of criteria to diagnose a definite concussion.

No repeated concussions or symptoms requiring further assistance were detected, but we consider that taking previous cases into account in future studies will probably help to increase the safety of the sports.

Active competition life in karate competition (an amateur sport) does not usually surpass 6–10 years, according to the entries recorded in the World Karate Championships (Accessed 30 November 2015.) Top karate athletes usually compete in one competition per month, with an average of four matches per competition (representing 12 min of active fighting in male categories and 8 min in female categories), or 144 min per year for men and 96 min for women. Extrapolating from the findings of our study (where a rate of definite concussion for male athletes was 1/9744 min of fighting, and in female athletes, it was 0/4200 min), we could consider that a male karate competitor has an exposure of 1440 min during his active competition life, with 2.2 suspected concussions on average during his athletic career, and 0.5 definite concussions. For a female karate competitor, exposure during her active competition life would be 960 min, with 1.8 suspected concussions on average, with a minimum risk of suffering a definite concussion. Taken globally, the risk of suffering a concussion while participating in high-level Olympic karate competition is low, making karate competition safe even when considered with other popular non-combat sports.14 In soccer, Boden et al24 found that the overall prevalence of college soccer-related concussions was 0.6 cases per 1000 AE for men and 0.4 cases per 1000 AE for women. We must also notice that in Olympic karate competition, the athletes do not use helmets as protection gear, as opposed to that used in Olympic taekwondo, but several studies have shown that the use of helmets in different sports has had no effect on the rate of concussions, and so the difference in concussion rate is most probably due more to the inherent characteristics of the sport and the competition rules than to the protective equipment used.25

One of the strengths of our study is the fact that data were collected prospectively, in situ, by medical personnel, and so the doubt on the honesty of the athletes to report the concussions that some other studies have taken into consideration may be eliminated,7 giving our results a robust consistency, and allowing for further comparison with other sports. Also, the fact that athletes suspected of being concussed were examined by highly trained medical personnel reduces the uncertainty about the diagnosis that other studies have pointed out as a possible bias.11 In the future, more studies analysing the risk of concussions in other competition levels of Olympic karate and different karate styles are needed, as they would help parents, coaches and doctors to make recommendations regarding the practice of safe sports. Nevertheless, it is important to note that, in spite of the high number of fights scrutinised, the number of concussions found in this study is very low, which makes statistical analysis difficult. From the results found in the present paper, we expect to expand the study of concussions in Olympic karate to other levels of competition (the recently created Premiere League competition, the different continental championships and even national level championships), and also different age frames (cadets, juniors and under 21 levels) in the future, using our data and methodology as a starting point of comparison.


The risk of concussions in WKF top-level karate competition is low, with a slightly increased risk for men and for competition without weight limits. It is not statistically significant with respect to women or individual competition and there is less than one expected concussion during a competition lifespan both for men and women.

What are the findings?

  • The rate of concussions in Olympic karate top-level competition is low.

  • The risk of concussions does not seem to be significantly different between men and women involved in Olympic karate top-level competition.

  • The risk of concussions does not seem to be significantly different in Olympic karate top-level competition for athletes involved in free-weight competition.

How might it impact on clinical practice in the future?

  • Recommendations for athletes involved in karate regarding the safety of the competition could be more precisely made, and recognition that concussions do occur in Olympic karate competition should help to promote a system for tracking the athletes suffering such injuries and protecting them.

  • Medical practicioners involved in the care of athletes participating in karate may have a starting point to counsel them in the case of concussions.


The authors thank Dr Benbekhma (Algerie), Dr Mlayah (Tunisia), Dr Orozco (Brazil), and Dr Zaeimkohan (Canada) and the medical teams working at the different championships for their tireless work recording the injuries that took place along the tournaments.



  • Twitter Follow David Inman @Davetheosteo and Miguel Saavedra @saavem

  • Contributors RA, FR, DC and DI conceived the project of the study and collected the data in situ. BA, MB, PR and MAS performed the bibliography research, analysed the results, helped with the statistical workup and critically read the manuscript, improving it. All authors approved the final version of the manuscript.

  • Competing interests None declared.

  • Ethics approval University of A Coruña Ethics Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.