An Analysis of Fencing Injuries in the United States: A 10-Year Database Review

Abstract

Background:

Fencing has maintained sustained popularity in the United States and internationally. However, there is limited information regarding acute injury patterns among fencing athletes.

Purpose:

To determine the prevalence of fencing injuries evaluated at emergency departments in the United States from 2013 to 2023 and analyze trends in diagnosis by demographics, mechanism of action, and disposition status.

Study Design:

Descriptive epidemiological study.

Methods:

The US Consumer Product Safety Commission’s National Electronic Injury Surveillance System (NEISS) database was queried in June of 2024 using the product code 3260–Fencing (activity/apparel/equipment). The analysis included descriptive statistics for diagnosis prevalence and demographics, univariate analysis utilizing diagnosis and demographic data, and a text analysis of the narrative descriptions.

Results:

From 2013 to 2023, 129 fencing-related injuries were reported from US emergency departments, representing an NEISS national estimate of 3418 (95% CI, 2200-4636). The most common age group affected was 11 to 15 years (35.7%), followed by 16 to 20 years (29.5%), with a median age of 16 years. Most injured individuals were White (49.65%) and male (56.5%). However, there was an upward trend in injuries among female athletes throughout the study. The most frequent injuries were strains or sprains (26.4%), lacerations (15.5%), fractures (11.6%), and contusions/abrasions (10.9%). Fracture prevalence was higher in females than males (14.4% vs 9.6%; P1 = .04). Males experienced a higher laceration prevalence (21.9% vs 7.1%; P = .02). The most affected body part was the finger (14%), followed by the wrist (8.5%), leg (8.5%), and head (7.8%). Most injuries occurred in sports-specific facilities (70%), followed by personal homes. Regarding disposition, 97.7% of all patients were treated and released. Fencing injuries rose steadily until the COVID-19 pandemic caused a decline, but they have since rebounded.

Conclusion:

Fencing injuries decreased significantly during the COVID-19 pandemic but have since become more prevalent. Fractures were more prevalent among females, whereas males had a higher prevalence of lacerations. Understanding the epidemiology and nature of fencing injuries can help inform coaches, athletes, and health care providers about potential risks and injury prevention strategies.

Fencing originated as a form of military training and has since evolved into a globally recognized sport, featured in every Olympics since the first modern games in 1896. ⁴ USA Fencing, the national governing body, boasts >40,000 members, with 44 US colleges offering National Collegiate Athletic Association (NCAA) sanctioned fencing programs, including 28 Division I schools.^14,15 USA Fencing’s 2024 to 2028 strategic plan aims to expand this growth further by adding 5 new NCAA programs by 2028. ¹⁸ This growth trajectory mirrors the sport’s expansion from 14,000 members in 1999 to >37,000 in the 2018-2019 season. ¹⁵ Despite setbacks during the COVID-19 pandemic, which saw membership drop to a 10-year low in January 2021, membership has rebounded to a record high in 2024.^5,15

Historically, fencing has been restricted to expensive private clubs and elite universities, with high start-up and operating costs. The sport has recently strived for a more diverse athlete base. Women’s épée debuted at the Olympics in 1996, followed by women’s saber in 2004, >100 years after men’s fencing. ⁴ Initiatives from the Peter Westbrook Foundation and Nzingha Prescod’s PISTE Academy have broadened access to the sport to traditionally underrepresented groups, including students from low-income families, Black and Latino students, and women.^6,9 Additionally, participation among veterans (>40 years of age) and para fencers has doubled from 2020 to 2024. ¹⁸ Despite fencing’s established presence in American and international sports, information on fencing-related injuries remains limited.

Fencing is divided into 3 weapons: foil, épée, and saber (Figure 1). Each weapon has its own set of rules and style of combat. The épée and foil are point weapons, where fencers use the tip or point of the blade to score. Saber fencers can use any part of the blade, earning points by either thrusting the tip forward or using the side of the blade in a cutting motion. Target areas vary: the épée targets the entire body, the foil targets the trunk, and the saber targets everything above the waist. Matches, known as bouts, take place on a 14 m–long and 1.5- to 2-m wide piste, with the main movement mechanics including the en garde position and lunge. ¹

Figure 1.

(A) The 3 weapons used in fencing: saber, foil, and épée. (B) Two fencers foil fencing and wearing fencing equipment, which includes knee-high socks, pants, a jacket, an underarm protector, an electronic vest for scoring (foil and saber), a plastic chest plate, a mask, and a glove.

The asymmetric en garde position consists of the weapon held forward, with the lead foot parallel, so both the blade and lead foot face the opponent. The rear foot is externally rotated. Fencers maintain stability in a quarter squat, using eccentric leg loading to initiate a lunge. ¹¹ The explosive lunge, starting with weapon extension, rapidly closes the distance to score. Recovery requires concentric force from the front foot to return to en garde. ¹³ Contact with the opponent is absorbed through the extended arm, and the cycle is repeated throughout the bout, posing unique injury risks. ¹³

The most severe fencing injury is a puncture wound to the internal organs, often in the setting of a broken blade or protective equipment failure. Enhanced protective equipment with greater coverage has largely mitigated these risks. ⁴ An example of this protective equipment can be seen in Figure 1. Previous work suggests that fencers’ lower limbs are most susceptible to injury, particularly the knee and hip joints. A prospective study in 2019 following college-level athletes found a high incidence of general musculoskeletal injuries, with the most common diagnoses being shoulder pain and ankle/foot sprain, and a higher prevalence of upper extremity injuries in men versus women (63% vs 33%). ¹⁶ Another study focusing on lower extremity injuries in US national and Olympic-level fencers ultimately found a higher number of knee injuries in the dominant (leading) knee, with the most common complaint being undiagnosed knee pain (37%) followed by meniscal injuries (9.8%). ¹² That study compared women and men collegiate athlete injuries and observed an increased number of hip injuries and lower levels of knee and hip function in women compared with their male counterparts. ¹² Previous work on fencing injury epidemiology has focused on chronic injuries through longitudinal data analysis.^8,12,16 To our knowledge, no previous work has been done exclusively on acute fencing injuries. The purpose of our study was to utilize the National Electronic Injury Surveillance System (NEISS) data set to identify the prevalence of fencing-related injuries in US emergency departments and analyze trends within demographics, mechanism of action, location, and disposition status of those patients. We hypothesized that given the target areas for all 3 weapons primarily involve the upper body, upper limb lacerations would be the most common contact injuries, while noncontact injuries would predominantly involve lower extremity strains and sprains due to fencing’s asymmetric biomechanics. Additionally, because men and women often compete and train together, we expected no significant sex differences in overall distribution or types of injuries sustained.

Methods

This study utilized the NEISS, which is updated by the US Consumer Product Safety Commission. The NEISS samples 100 emergency departments with at least 6 beds in the United States and its territories. The sample is stratified by hospital size and location to ensure a wide geographic spread that is ultimately used as a probability sample of all hospitals meeting inclusion criteria. A national estimate of injuries by product type can be calculated using NEISS-assigned weights for each case. Because this database is publicly available and de-identified, this study was exempt from institutional review board approval.

The NEISS database is populated by participating hospitals, which use a standardized guidebook of product codes to accurately document the specific products or activities associated with each injury. For this study, the database was queried from 2013 to 2023 using the product code 3260–Fencing (activity/apparel/equipment). The query resulted in a list of computer-generated identification numbers representing the case, along with case information. The case information included injury type, age group, patient race, body part injured, injury setting, primary and secondary diagnoses, disposition after the emergency department presentation, and a short narrative.

Analysis by Age Group and Sex

Cases were reclassified into age groups and used in a univariate analysis to examine the most common diagnoses by age group. A similar method was applied to sex data to determine trends in diagnosis by sex along with prevalence of injury. A subgroup analysis comparing pediatric athletes (<18 years) and veteran division athletes (>40 years) was also completed.

Analysis by Mechanism, Setting, and Weapon

Fencing-related narrative texts and the location of injury were utilized to elucidate the mechanism of injury and setting (practice or tournament). This was performed with input from coauthors B.S., J.X.C., and A.L.S. The mechanism of injury was assigned one of the following categories: fall, hit/strike, noncontact musculoskeletal injury, and other. Weapons were assigned, when available, into the 3 fencing weapons: épée, foil, and saber. If the weapon type was not specified, the entry was labeled as “unknown.” Univariate analysis was done to determine the diagnoses associated with each mechanism of injury, injury setting, and fencing weapon.

Statistical Analysis

Descriptive statistics were performed for diagnosis prevalence and patient characteristics. Categorical variables were compared using cross-tabulations and analyzed using chi-square or Fisher exact tests. Subgroup analyses of diagnoses were done based on patient sex, age group, setting, and mechanism of injury. All statistical tests were performed using IBM SPSS Version 29 (IBM Corp) and visualized using both SPSS and Microsoft Excel. For all analyses, statistical significance was set at a P value <.05.

Results

During the study period, 129 fencing-related injuries were evaluated at US emergency departments, representing an NEISS national estimate of 3418 (95% CI, 2200-4636). There was not a linear annual trend in injury prevalence, with injuries remaining stable from 2014 to 2019. Between 2019 and 2022, injuries reached a low, with only 5.4% of all injuries occurring in 2022. The data peaked in 2023, representing 14.7% of all injuries. Trends in total injury numbers over time are displayed in Figure 2. White patients were the largest group represented (49.6%), and the prevalence in Asian athletes has steadily increased since 2020.

Figure 2.

Number of fencing-related injuries by year (2013-2023).

The median patient age was 16 years (mean, 22 years; range, 4-68 years). The most common age group was 11 to 15 years, representing 35.7% of all injuries, followed by 16 to 20 years, representing 29.5% of all injuries. Fractures were more likely to be sustained by those 11 to 15 years of age (40%) but also represented the most common injury pattern in the oldest age group (66-70 years). The most common pediatric (<18 years) injury was a strain or sprain (n = 16), which were 50% upper extremity and 50% lower extremity. The most common injury in veteran division fencing (>40 years) was lower extremity unspecified pain. Patients aged 16 to 20 years sustained the most concussions (66.7%).

Nearly all patients were treated and released (97.7%). Two patients were admitted, one for a pediatric femur fracture after a fall, and the other, a veteran division fencer, for a non–ST segment elevation myocardial infarction. One fencer, diagnosed with rhabdomyolysis that was linked to his fencing gear during a tournament, was held for observation.

The most common diagnosis was a strain or a sprain, followed by laceration and fracture, as depicted in Table 1. The fracture rate was higher among female patients (14.3% vs 9.6%; P = .04), as was the concussion rate (7.1% vs 2.7%; P = .06), but this was not significant. Males experienced significantly more lacerations (21.9% vs 7.1%; P = .02) and contusions/abrasions (15.1% vs 5.4%; P = .04). A comparison of number of diagnoses by sex is depicted in Figure 3, and the course of diagnosis by sex is depicted in Figure 4.

Table 1.

Fencing-Related Injury Diagnosis Frequency and Percentage of Total

Diagnosis	Frequency	Percent
Concussion	6	4.7
Contusion, abrasion	14	10.9
Dermatitis, conjunctivitis	1	0.8
Dislocation	5	3.9
Foreign body	2	1.6
Fracture	15	11.6
Internal organ injury	2	1.6
Laceration	20	15.5
Nerve damage	1	0.8
Other/not stated	24	18.6
Puncture	5	3.9
Strain or sprain	34	26.4

Figure 3.

Ten-year diagnosis percentage of fencing-related injuries by sex.

Figure 4.

Fencing-related injury prevalence over 10 years separated by sex.

Most fractures occurred in the finger, wrist, and leg, with each accounting for 20% of all fractures. Lacerations were most frequently observed in the finger (35%), followed by the face (20%). Strains and sprains were most commonly reported in the ankle (25.5%) and finger (17.6%). Diagnoses by body location are presented in Table 2. Percentages reflect the proportion of each diagnosis within the full distribution for that body part. Only the three most common diagnoses per body part are displayed; less frequent diagnoses are excluded from the table.

Table 2.

Most Common Acute Diagnoses by Body Part in Fencing Injuries

Diagnosis	Frequency	Percent
Finger (n = 18)
Laceration	7	38.9
Strain/sprain	6	33.3
Fracture	3	16.7
Wrist (n = 11)
Strain/sprain	4	36.4
Fracture	3	27.3
Dislocation	1	9.1
Leg (upper and lower) (n = 11)
Strain/sprain	4	36.4
Fracture	2	18.2
Laceration	2	18.2
Head (n = 10)
Concussion	6	60.0
Contusion/abrasion	1	10.0
Laceration	1	10.0
Trunk (upper and lower) (n = 13)
Unspecified pain	10	76.9
Fracture (clavicle)	2	15.4
Strain/sprain	1	7.7

Table 3 describes the location where the injury occurred. The majority of injuries took place at sport-specific venues (70%), followed by home (5.4%). Eighteen percent of all injuries occurred during practice, with 6 related to being hit with a blade, 1 resulting from a fall, and 4 due to noncontact musculoskeletal injuries. Although only 1 fall was recorded during practice, it resulted in a fracture, with the narrative describing a fall on an outstretched hand. Nine percent of all injuries were reported during tournaments, and 25% of those were diagnosed as sprains. The level of tournament competition was not included in the narrative descriptions. Two injuries were specifically attributed to foil and 5 to saber. All weapon-specific injuries involved a hit from a blade, with no significant patterns in injury type by weapon. No épée-related injuries were specifically recorded.

Table 3.

Fencing-Related Injuries Occurring in Practice and Tournament Settings

	Frequency	Percent
Injuries in tournament settings (n = 23)
Hit/strike	8	34.8
Musculoskeletal	5	21.7
Fall	8	34.8
Syncope	2	8.7
Injuries in practice settings (n = 13)
Hit/strike	6	46.2
Musculoskeletal	4	30.8
Fall	1	7.7
Other	2	15.4

Discussion

The findings of our study revealed that fencing injuries plateaued from 2014 to 2019; experienced a sharp decline between 2019 and 2022 to 5.4% in 2022, correlating with COVID-19; and started rebounding in 2023, peaking at 14.7%. This pattern reflects the state of fencing nationally in the past decade, which was experiencing steady growth until COVID-19 caused a significant setback before returning to its original pace of athlete growth.^5,15 Demographically, the majority of injuries occurred in males (56.5%), largely between 11 and 15 years of age (35.7%), and the most common race represented was White (49.6%). However, in examining the data, other trends emerge, including an increase in injury prevalence in women along with a steady increase in injuries among Asian athletes.

In this study, sex played a significant role in diagnoses, with women having a significantly higher fracture rate than men (14.4% vs 9.6%; P = .04). This supports other epidemiological papers on fencing injuries, including a previous study that found a 2.6 times incidence in females, although this current study only found a 1.5 times incidence. ¹⁶ Previous work on fracture rates between male and female athletes is largely focused on stress fractures, with studies supporting a higher stress fracture rate in females. ¹⁰ However, data on nonstress fractures in pediatric and adolescent populations are more limited. Many papers on sports-related fractures have a male majority in their study population, and work on fall-related fractures has largely been done in the more vulnerable elderly population. One study on athlete fracture rates from several sports found an increased rate of radius fractures in males, while females had a higher rate of femoral fractures. ¹⁷ Our data did not support this, with most female fractures located in the upper extremity (63%), including the wrist, clavicle, and finger, while males had a more variable fracture pattern but experienced only 2 leg fractures.

In our study, females had a higher concussion rate (7.1% vs 2.7%; P = .06); however, this was not significant, potentially due to the limited sample size. This overall trend supports work done by the NCAA Injury Surveillance Program, which found that in sex-comparable sports, females had a 1.4 times higher overall concussion risk than males. ² With the increasing sex, ethnic, and age diversity in the sport of fencing and low representation in the existing literature, there is a need to examine the changing population for differences in injury risk to inform future safety measures and ensure the applicability of findings to the athletes.

Significant lacerations and puncture wounds are relatively rare in fencing despite the sport’s origin in combat but were surprisingly significant in this data set. Low overall laceration and puncture risk in fencing as a whole can be attributed to highly advanced fencing equipment made of double-thickness Kevlar or cotton and wire mesh masks. ⁴ Historically, the fencing community has been quick to respond to severe injuries associated with lacerations and punctures, with safety improvements including an underarm plastron around the leading arm/weapon-side chest that doubles the amount of protective fabric in the area closest to a fencer’s opponent. ⁴ Nevertheless, the laceration rate was high in this study and was particularly higher in men (21.9% vs 7.1%; P = .02). Many of the lacerations in this study were associated with body parts that are entirely covered by regulation fencing equipment, including the face, neck, eye, and upper thigh. This may indicate that some athletes, particularly young males, may be practicing without proper gear, leading to an increased risk of injury.

Proper gear, however, is not without its risks. One patient in our study, a 17-year-old male, was in full gear when he developed leg cramps, which turned out to be rhabdomyolysis. The thick protective clothing can affect the body’s ability to dissipate heat effectively, potentially increasing the risk for complications, including heat stroke and rhabdomyolysis. ¹

Age group also played a significant role in diagnosis, with the most fractures occurring in the 11- to 15-year age group (40%). However, fractures were also the most common injury pattern in the oldest age group (66-70 years), which is concerning given the injury severity of a fracture and the decreased healing ability of older athletes. That said, the most common injury in veteran division fencing (>40 years) was unspecified pain, supporting evidence that fencing is a relatively safe sport for older athletes. Little previous work has been done in older or veteran division combat sports, with 1 study in boxing finding that 40- to 59-year-old patients had a higher proportion of shoulder sprains and hand fractures than other age groups. ⁷

The overall safety of fencing is supported by the disposition of patients, as nearly all patients were treated and released without the need for hospitalization. This conclusion is also supported by high rates of strains/sprains and generalized pain, as opposed to more serious time-loss injuries such as fractures and dislocations, consistent with previous literature. ³ Additionally, finger injuries, a relatively minor yet inconvenient injury, were more common than the more life-threatening leg, arm, and head injuries. Unfortunately, although some NEISS narratives provided laterality, it is impossible to know if injuries occurred on fencers’ dominant or nondominant (weapon-wielding or nonweapon) sides, which could change the interpretation of the data and provide insight into the effects of asymmetry. Previous work suggests increased risk to fencers’ dominant side, including a higher rate of dominant knee and hip injuries. ¹²

Nearly all injuries occurred in a sports-specific venue, followed by home. This location pattern makes sense given the equipment demands, including the specialized flooring (aluminum or conductive fabric) of which fencing pistes are composed, and electronic components necessary for accurate scoring. In total, 36 cases (28% of all cases) specified the level of competition in which the injury occurred, whether it be a practice or tournament. Eighteen percent of all injuries occurred at practice, with an equal prevalence of hit/strike injuries and falls. An interesting statistic showed that 75% of falls during practice resulted in a fracture, most of which were falls on an outstretched hand. This may be linked to the hard piste surface or the challenge of catching oneself after falling with 1 hand gripping a fencing blade. Little recording was done to specify the style of fencing involved in the injuries, possibly because of a general lack of knowledge surrounding fencing. Five percent of injuries noted the weapon type involved. Supporting previous work, saber fencing had the most injuries, followed by foil, although there was no mention of épée injuries. ¹² All weapon-specific injuries were related to a hit from the blade, but there were no identifiable trends in the injury patterns sustained from those hits.

Limitations

While this study aimed to add to the limited body of work representing fencing injuries, there are several limitations, many of them stemming from the NEISS database. The NEISS database is limited by the information that providers and research assistants decide to input, with narratives ranging in length and level of detail. There were few variables per case, and having additional information, including time spent in the hospital, diagnostic testing, and surgical/nonsurgical interventions, would help us further understand how injuries were managed. Although there is no information about treatment or long-term outcomes in the NEISS database, disposition data were valuable in inferring injury severity. Additionally, the NEISS database represented <4% of the fencing injuries that were predicted to have occurred during the study period. This relatively limited study population may be linked to the inclusion of only emergency departments, with other acute injuries ending up at urgent care centers or other health care settings, remaining unexamined. It also may be explained by the geography of fencing participation, which is mostly concentrated at the coasts, with the highest participation rates from New Jersey, New York, and California, while NEISS is geared toward nationwide catchment. This limits the generalizability of our results. Finally, the NEISS database, while a valuable tool for sports injury research, currently lacks comprehensive coverage of combat sports beyond boxing and mixed martial arts. This limitation is particularly significant for sports like fencing and kendo, which, despite sharing some similarities with other combat sports, have unique rules, equipment, and injury patterns. The absence of significant data from these sports hinders our ability to conduct a comprehensive comparative analysis. Despite these limitations, this study is a valuable addition to the literature on acute fencing injuries.

Conclusion

Between 2013 and 2023, trends in fencing injuries reflected those trends within the sport of fencing overall, including the broadened diversity in race, sex, and age group of fencing athletes. The most common injury was a strain or a sprain, followed by a laceration. The most common body parts impacted were the finger, wrist, and leg. Although there was male predominance in the data set, females experienced a higher fracture, while males experienced a higher laceration rate. Understanding the prevalence and nature of fencing injuries can help inform coaches, athletes, and health care providers about potential risks and injury prevention strategies. Further work is needed to elucidate the nature of acute injuries in fencing, particularly utilizing data that include weapon type and handedness, which could ultimately be used to make the sport even safer than it is, allowing it to continue its growth.