Part II: Comparison of Crossfit-Related Injury Presenting to Sports Medicine Clinic by Sex and Age

Abstract

Objective:

To examine CrossFit-related injuries based on sex and age.

Design:

Retrospective case series.

Setting:

A tertiary-level pediatric sports medicine clinic.

Participants:

CrossFit athletes.

Main Outcome Measures:

CrossFit-related injuries by sex (males vs females) and age groups (≤19 years vs >19 years) using a χ² analysis with P = 0.05, odds ratio (OR), and 95% confidence interval (95% CI).

Results:

Among injured CrossFit athletes, female athletes sustained lower extremity injuries more frequently than male athletes (P = 0.011; OR, 2.65; 95% CI, 1.25-5.65). In observed CrossFit injuries, shoulder injuries were more frequently observed in male athletes compared with female athletes (P = 0.049; OR, 2.79; 95% CI, 0.98-7.95). Additionally, a greater proportion of CrossFit athletes aged 19 years and younger suffered trunk/spine injuries than those older than 19 years (P = 0.027; OR, 2.61; 95% CI, 1.10-6.21) in injured CrossFit athletes.

Conclusions:

The current results indicated sex- and age-specific susceptibility to CrossFit-related injuries based on body parts and diagnoses. The presented information may be useful to develop a safer exercise program, especially for pediatric and adolescent CrossFit participants.

INTRODUCTION

CrossFit is a form of exercise that incorporates a combination of Olympic-style weight lifting, powerlifting, high-intensity interval training, and plyometric movements.^1-3 Participation in CrossFit can include a variety of benefits including improving patellar and Achilles tendon thickness.⁴ Additionally, a randomized controlled study of CrossFit Teens reported numerous health- and fitness-related benefits, including improvement in waist circumference, body mass index (BMI)-Z scores, sit and reach, standing jump, and shuttle run.⁵ Although the positive effects of CrossFit participation have been reported, there are limited data regarding relative injury risks in youth CrossFit participation. Prior reports indicated that approximately 20% of adult CrossFit athletes sustain injury.² Injury rates occur at 3.1 cases per 1000 hours trained,⁶ and the most frequently injured body parts were shoulder, knee, and low back in adult CrossFit athletes.⁷

Although epidemiological studies on CrossFit-related injuries are documented, the population of interest in these investigations is predominantly adults. To our knowledge, there are no existent studies that focus on examining CrossFit injury incidence in pediatric and adolescent populations. The relative injury risk of CrossFit programs tailored for children and teenagers is unknown because of the lack of data on injury epidemiology in this group. To optimize safe participation of youth in CrossFit, there needs to be an increased understanding of common injury patterns in this population. Furthermore, pediatric athletes sustain different types of injuries than adolescent athletes.^8-10 In addition, injury profiles in youth athletes can differ by sex.⁹ Evidence that compares CrossFit injuries between sex and age is notably limited. Therefore, this study has 2 primary aims: to examine CrossFit-related injuries between male and female athletes and to compare CrossFit-related injuries in athletes between 19 years and younger and older than 19 years.

METHODS

Data and Study Participants

A single-center chart review was conducted at a tertiary-level pediatric sports medicine clinic between January 1, 2003, and June 31, 2016. The Committee on Clinical Investigation of the Hospital Institutional Review Board approved the study protocol before commencement. Study participants were selected if the term “crossfit” was identified using a departmental electronic medical records database. Inclusion criteria for this study encompassed all injuries that were deemed a direct result of CrossFit participation by the athlete and previous injuries that were exacerbated by CrossFit participation. We defined injury as a change in the current state of health to a less-healthy state as a direct result of a CrossFit workout. Patients were excluded from analysis if injuries were not related to CrossFit, a direct association between the injury and CrossFit was not able to be determined or if the patient had medical conditions that could influence their risk for injury outside of CrossFit (cerebral palsy, Ehlers-Danlos, Chiari malformation, Lyme disease, or hemiplegia).

All outpatient clinic notes identified as CrossFit cases were reviewed by the leading research team member (R.L.Z) to ensure that the mechanism of injury occurred as a result of participation in CrossFit and that all inclusion and exclusion criteria were applied. According to institutional standards, all clinic notes had been reviewed and approved by a fellowship-trained attending physician or surgeon within the Department of Orthopaedic Surgery and/or Division of Sports Medicine.

Main Outcome Measures

Data collection included date of injury, demographic information (sex and age), and injury profiles (injured body sites, diagnosis). Main outcome measures were injured body sites and diagnoses, which were compared by sex and age.

Statistical Analysis

Patient demographic information, including age, height, weight, and BMI, were analyzed by descriptive statistics including mean and SD. Frequencies (N) of CrossFit-related injuries were expressed as percentages (%), and the results were calculated by developmental stages (≤19 years and >19 years) and sex (males and females) separately. Commonly injured body parts were categorized by the following body locations: head, upper extremity, trunk/spine, and lower extremity. Chi-square (χ²) analysis was used to examine proportional differences of CrossFit-related injuries to the head, upper extremity, trunk/spine, and lower extremity based on developmental stages (≤19 years and >19 years) and sex (males and females). The 4 body locations were further broken down into specific body parts/joints: foot, ankle, knee, hip, spine, shoulder, elbow, and wrist. Injured body parts/joints that were found more than 10 times were analyzed by χ² analysis to determine proportional differences by developmental stages (≤19 years vs >19 years) and sex (males vs females). Along with statistical significance of P ≤ 0.05, odds ratio (OR) with 95% confidence interval (95% CI) was also analyzed. Furthermore, whether BMI plays a significant role in specific injured body parts/joints was tested by Student t test when data were normally distributed. For non-normal data, Mann-Whitney U test was used. The priori statistical significance was set as P ≤ 0.05. All analyses were performed using SPSS statistical software (Version 23, SPSS, Inc, Chicago, IL).

RESULTS

One-hundred fifteen medical charts met the inclusion criteria for CrossFit injuries. Demographics of CrossFit athletes were displayed by sex; male (N = 55) and female (N = 60) (Table 1). Similarly, physical characteristics of athletes aged 19 years and younger (N = 40) and those older than 19 years (N = 75) were illustrated (Table 2). Specific diagnoses of CrossFit-related injuries were listed by sex in Tables 3 and 4. A list of specific diagnoses by age 19 years and younger and older than 19 years was displayed in Tables 5 and 6. Because there was only 1 head injury, χ² analysis was performed based on 3 major body regions: upper extremity, trunk/spine, and lower extremity. Also, knee, spine, shoulder, and hip/pelvis were found more than 10 times in the chart review; thus, χ² analysis was used to compare the effect of sex and developmental stage for each of these body parts/joints. Chi-square was usedto assess the effect of sex and developmental stages. For BMI analysis, Shapiro-Wilk test indicated non-normal data distribution. Thus, Mann-Whitney U test was applied to determine whether there are BMI differences in rates of injury to knee, spine, shoulder, and hip/pelvis joints.

TABLE 1.

Physical Characteristics of Male and Female CrossFit Participants

Variables	Males (n = 55)	Females (n = 60)
Age (yrs)	24.4 ± 10.8	26.0 ± 10.1
Height (cm)	173.9 ± 10.7	164.8 ± 7.6
Weight (kg)	77.1 ± 17.2	65.7 ± 14.4
BMI	25.2 ± 4.4	24.3 ± 5.4

TABLE 2.

Physical Characteristics of Adolescent and Adult CrossFit Participants

Variables	≤19 Years (n = 40)	>19 Years (n = 75)
Age (yrs)	15.2 ± 2.5	30.6 ± 8.9
Height (cm)	165.7 ± 10.5	170.9 ± 9.7
Weight (kg)	64.8 ± 16.5	74.6 ± 15.9
BMI	23.5 ± 5.2	25.4 ± 4.6

TABLE 3.

A List of Diagnoses—Male

Diagnosis	Frequency (N)	Percentages (%)
Generalized back pain	4	7.5
Disc herniation: L5-S1	4	7.5
Muscle strain	3	5.7
Generalized knee pain	3	5.7
Patellar tendinitis	3	5.7
Spondylolysis	3	5.7
Contusion	2	3.8
Tendonitis	2	3.8
Meniscus tear	2	3.8
Disc protrusion	2	3.8
Ligament sprain	1	1.9
Degenerative joint irritation	1	1.9
Back strain	1	1.9
Rotator cuff impingement	1	1.9
Femoral acetabular impingement	1	1.9
Bicep tendonitis	1	1.9
Fracture: wrist	1	1.9
Generalized arm pain	1	1.9
Patella subluxation	1	1.9
Generalized leg pain	1	1.9
Glenohumeral internal rotation irritation	1	1.9
Labral tear	1	1.9
Sacroiliac joint injury	1	1.9
Hypoplasia elbow	1	1.9
Concussion	1	1.9
Sternoclavicular joint instability	1	1.9
Iliotibial band syndrome	1	1.9
Epiocondylitis	1	1.9
Osteochondritis dissecans	1	1.9
Chondromalacia	1	1.9
Generalized shoulder pain	1	1.9
Tendon rupture	1	1.9
AC joint injury	1	1.9
Sever disease	1	1.9
Generalized rib pain	1	1.9

TABLE 4.

A List of Diagnoses—Female

Diagnosis	Frequency (N)	Percentages (%)
Patellofemoral stress syndrome	5	8.3
Muscle strain	4	6.7
Labral tear	4	6.7
Bursitis	3	5.0
Sacroiliac joint injury	3	5.0
Biceps tendonitis	2	3.3
Tendonitis	2	3.3
Generalized back pain	2	3.3
Generalized ankle pain	2	3.3
Ligament tear	2	3.3
Meniscus tear	2	3.3
Quad tendinopathy	2	3.3
Generalized knee pain	2	3.3
Generalized foot pain	2	3.3
Generalized hip pain	2	3.3
Spondylolysis	1	1.7
Back strain	1	1.7
Rotator cuff impingement	1	1.7
Tibial sesamoiditis	1	1.7
Femoral acetabular impingement	1	1.7
Stress fracture: spine	1	1.7
Rotator cuff tendonitis	1	1.7
Fracture: wrist	1	1.7
Bilateral L5 pars defect	1	1.7
Generalized arm pain	1	1.7
Iliac crest apophysitis	1	1.7
L5-S1 disk bulge	1	1.7
Generalized leg pain	1	1.7
Stress fracture: knee	1	1.7
Supine impingement	1	1.7
IT band syndrome	1	1.7
Ruptured baker cyst	1	1.7
Epicondylitis	1	1.7
Stress fracture: hip	1	1.7
Periostitis	1	1.7

TABLE 5.

A List of Diagnoses—≤19 Years

Diagnosis	Frequency (N)	Percentages (%)
Generalized back pain	4	10.3
Spondylolysis	4	10.3
Patellar tendinitis	3	7.7
Patellofemoral stress syndrome	3	7.7
Sacroiliac joint injury	3	7.7
Back strain	1	2.6
Muscle strain	1	2.6
Overuse of joint	1	2.6
Rotator cuff impingement	1	2.6
Tibial sesamoiditis	1	2.6
Contusion	1	2.6
Femoral acetabular impingement	1	2.6
Degenerative joint irritation	1	2.6
Rotator cuff tendonitis	1	2.6
Fracture: wrist	1	2.6
Bilateral L5 pars defect	1	2.6
Generalized arm pain	1	2.6
Generalized ankle pain	1	2.6
Iliac crest apophysitis	1	2.6
Patella subluxation	1	2.6
Tendonitis	1	2.6
Glenohumeral internal rotation irritation	1	2.6
Hypoplasia elbow	1	2.6
Concussion	1	2.6
Generalized knee pain	1	2.6
Stress fracture: knee	1	2.6
Sever disease	1	2.6

TABLE 6.

A List of Diagnoses—>19 Years

Diagnosis	Frequency (N)	Percentages (%)
Muscle strain	6	8.1
Labral tear	5	6.8
Disc herination: L5-S1	4	5.4
Meniscus tear	4	5.4
Tendonitis	4	5.4
Generalized knee pain	4	5.4
Biceps tendonitis	3	4.1
Patellofemoral stress syndrome	2	2.7
Generalized foot pain	2	2.7
Bursitis	2	2.7
Generalized back pain	2	2.7
Ligament tear	2	2.7
Iliotibial band syndrome	2	2.7
Generalized hip pain	2	2.7
Quad tendinopathy	2	2.7
Generalized leg pain	2	2.7
Disc protrusion	2	2.7
Epicondylitis	2	2.7
Rotator cuff impingement	1	1.4
Contusion	1	1.4
Femoral acetabular impingement	1	1.4
Stress fracture: spine	1	1.4
Fracture: wrist	1	1.4
Back strain	1	1.4
Generalized arm pain	1	1.4
Generalized ankle pain	1	1.4
L5-S1 disc bulge	1	1.4
Ligament sprain	1	1.4
Sacroiliac joint injury	1	1.4
Supine impingement	1	1.4
Sternoclavicular joint instability	1	1.4
Ruptured baker’s cyst	1	1.4
Osteochondritis dissecans	1	1.4
Chondromalacia	1	1.4
Generalized shoulder pain	1	1.4
Stress fracture: hip	1	1.4
Tendon rupture	1	1.4
Periostitis	1	1.4
Acromioclavicular joint injury	1	1.4
Generalized rib pain	1	1.4

Injury Diagnoses by Sex

There were no proportional differences in upper extremity injuries (male: 18/55, 32.7%; female: 13/60, 21.7%; P = 0.182) or in trunk/spine injuries (male: 17/55, 30.9%; female: 12/60, 20.0%; P = 0.178) by sex. However, there was a difference in lower extremity injury by sex; χ² analysis showed that there was a greater proportion of females with lower extremity injury than males (male: 19/55, 34.5%; female: 35/60, 58.3%; P = 0.011; OR: 2.65; 95% CI, 1.25-5.65). With regards to the specific body parts/joints, there were no differences in knee (male: 13/55, 23.6%; female: 18/60, 30.0%; P = 0.442), spine (male: 16/55, 47.8%; female: 12/60, 20.0%; P = 0.257), and hip/pelvis (male: 3/55, 5.5%; female: 8/60, 13.3%; P = 0.151) injuries between male and female athletes. There was a statistically significant difference involving the shoulder joint with male athletes having greater CrossFit-related injuries than female athletes (male: 13/55, 23.6%; female: 6/60, 10.0%; P = 0.049; OR: 2.79, 95% CI, 0.98-7.95).

Injury Diagnoses by Age

There were no proportional differences in upper extremity (≤19 years: 7/40, 17.5%; >19 years: 24/75, 32.0%; P = 0.095) and lower extremity (≤19 years: 17/40, 42.5%; >19 years: 37/75, 49.3%; P = 0.484) injuries between athletes aged 19 years and younger and those older than 19 years. However, χ² analysis indicated greater trunk/spine CrossFit injury proportions in athletes aged 19 years and younger than in those older than 19 years (≤19 years: 15/40, 37.5%; >19 years: 14/75, 18.7%; P = 0.027; OR, 2.61; 95% CI, 1.10-6.21). Chi-square analysis was performed for body parts/joints recorded more than 10 times, which include knee, hip/pelvis, shoulder, and spine. In the analysis, there were no statistical differences in knee (≤19 years: 11/40, 27.5%; >19 years: 20/75, 26.7%; P = 0.924), hip/pelvis (≤19 years: 2/40, 5.5%; >19 years: 9/75, 12.0%; P = 0.224), or shoulder (≤19 years: 4/40, 10.0%; >19 years: 15/75, 20.0%; P = 0.169) injury proportions between CrossFit athletes aged 19 years and younger and those older than 19 years. However, a statistically significant difference was found when analyzing injuries to the spine. There were greater CrossFit-related spine injuries in those aged 19 years and younger than those older than 19 years (≤19 years: 15/40, 37.5%; >19 years: 13/75, 17.7%; P = 0.016; OR, 2.86; 95% CI, 1.19-6.87).

Body Mass Index and Knee, Spine, Shoulder, and Hip/Pelvis Joints Injury Status

There were no statistically significant BMI differences based on the status of knee (CrossFit athletes with knee injuries: BMI, 24.3 ± 4.7; CrossFit athletes without knee injuries: BMI, 24.9 ± 5.0; P = 0.405), spine (CrossFit athletes with spine injuries: BMI, 24.9 ± 4.2; CrossFit athletes without spine injuries: BMI, 24.7 ± 5.1; P = 0.552), shoulder (CrossFit athletes with shoulder injuries: BMI, 25.0 ± 4.2; CrossFit athletes without shoulder injuries: BMI, 24.7 ± 5.1; P = 0.362), and hip/pelvis (CrossFit athletes with hip/pelvis injuries: BMI, 24.0 ± 2.3; CrossFit athletes without hip/pelvis injuries: BMI, 24.8 ± 5.1; P = 0.840) injuries.

DISCUSSION

In this study, we sought to investigate injuries sustained by CrossFit and compare these injuries in younger versus older participants and males versus female athletes. One of the main findings of this current investigation was that CrossFit athletes aged 19 years and younger had a greater proportion of spine-related injuries than those older than 19 years. The most common spine-related injuries in the 19 years and younger group included generalized back pain, sacroiliac joint injury, and spondylolysis (Tables 5 and 6). Several studies reported the lower back as one of the most commonly injured body parts in CrossFit participants.^2,7,11 Although the spine was documented as a common location for CrossFit injuries, susceptibility of youth CrossFit athletes to spine-related injuries has not been identified. According to a study performed by Weissenthal et al,² the spine is most often injured during powerlifting maneuvers in adult men and women.² The commonly used power lifting movements in CrossFit are squats and deadlifts. A combination of substantial weight and inadequate techniques may be contributors to back pain and sacroiliac joint injuries (Table 5 and Table 6). Spondylolysis is prevalent among athletes who participate in sports that incorporate repetitive hyperextension of the lower back such as throwers, gymnasts, and rowers.¹² Therefore, a potential mechanism for spondylolysis in youth CrossFit athletes may be Olympic and/or powerlifting maneuvers with inadequate core/trunk stability and excessive load of the posterior spinal regions. Spine injuries, specifically, spondylolysis, are usually treated with nonsurgical management.¹³ However, if surgical intervention is necessary, fusion or direct pars repair of the spine is often performed.¹³ Even though surgical intervention may be successful, most of the patients require several months of rehabilitation.¹⁴ Additionally, according to Radcliff et al,¹⁴ a substantial number of those who had surgical intervention do not reach to their preinjury level of activity. Therefore, injury prevention strategies in youth CrossFit athletes need further investigation and implementation. Poor supervision was identified as a possible contributor to injury in prior reports evaluating strength training in youth.¹⁵ The majority of youth resistance training injuries are the result of accidents that are potentially preventable with increased supervision and stricter safety guidelines.¹⁶ Prior reports indicate reduced injury rates when a trainer was present during CrossFit training.² Hence, specific instructions and appropriate progressions need to be considered for young CrossFit athletes.

Another finding of our study revealed more frequent shoulder joint injuries in males than in females among injured CrossFit athletes. Shoulder-related injuries often occur with gymnastic-type movements.² According to the CrossFit theoretical hierarchy of development,¹⁷ the gymnastic movement was incorporated to establish functional capacity for body control and range of motion. Establishing the body control and range motion is theorized to further assist developing ability to control external objects and generate power through weightlifting. All of those components are programmed to facilitate various sport movements and skill sets. Those concepts are unique and differentiate CrossFit from other forms of weight training. The combination of the extreme technical requirements of the motion and equipment, in the setting of minimal prior training, may be one part of the explanations for greater proportion of shoulder injuries observed in male CrossFit athletes than female CrossFit athletes. Also, Olympic weightlifting places significant demand on the shoulder via overhead weighted maneuvers, such as the snatch. Furthermore, Montalvo et al⁷ reported that greater height and mass (body weight) are related to CrossFit injuries. Therefore, more frequent shoulder injuries found in males relative to females may be reflective of size differences. Prior CrossFit-related studies indicated that the shoulder is the most frequently injured body part in CrossFit athletes.^2,7,11,18 Our data showed statistical significance, albeit marginally significant (P = 0.049). Thus, more studies are needed to establish an association among body height and mass, sex differences, and CrossFit injury risk.

Another finding in our study includes a greater proportion of lower extremity injuries detected in female CrossFit athletes when compared with male athletes. Previous reports have supported these results. During strength training activities, women demonstrated a higher risk of accidental and lower extremity injuries more frequently than men. Conversely, men suffered more exertional type resistance training injuries, such as sprains and strains, compared with women, particularly in the trunk.¹⁹ Additionally, past studies have shown the knee joint as one of the more frequently injured joints in CrossFit.^2,7 This prompted us to evaluate knee joint–related CrossFit injuries between male and female athletes. We found, however, no statistical differences seen for injuries involving the knee joint (male: 13/55, 23.6%; female:18/60, 30.0%; P = 0.442). Similarly, there were no differences in hip/pelvis (male: 3/55, 5.5%; female: 8/60, 13.3%; P = 0.151). Although statistical significance was not found, percentages (%) were greater in female athletes than in male athletes. It is interesting to note that there were 5 cases of patellofemoral stress syndrome (PFSS) in female athletes (Tables 3 and 4). Repetitive stress and high ground impact forces seem to contribute to the development of PFSS.²⁰ A recent study found that females with PFSS have greater knee valgus while performing a single-leg squat and landing maneuvers.^21-23 Other studies found greater knee valgus angles with individuals with weak hip abductors.^24,25 Therefore, female CrossFit athletes who had weak hip abductors executing complex weightlifting maneuvers to load and/or time, repetitively jumping (box jumps, jump rope) might have an increased propensity to develop PFSS. Furthermore, increased injury rates have been associated with high exposure to training volume and length of participation.⁷ To test this hypothesis, future studies are warranted to conduct a laboratory-controlled study to clarify the association among training volume, muscular strength, and injury rate.

Although there are greater risks to certain body locations and specific body parts/joints based on sex and developmental stages, CrossFit participation confers many benefits, including muscle hypertrophy, body composition, and fitness abilities.^4,5 Integrating strength and conditioning programs into training regimens has been theorized to optimize future athletic development.^26,27 There is a specialty CrossFit course for kids called “CrossFit Kids program.” The CrossFit Kids program provides instructors with additional information on promotion of life-long fitness and adaptation of training routines for children and adolescents. Furthermore, several articles suggest that as long as qualified instructors and appropriate supervision are provided, strength and conditioning program may possibly reduce athletically related injuries and pathologies in physically active youth.²⁸ Several studies discussed the benefits and risks of CrossFit training^29,30; however, it is important to longitudinally investigate long-term effects of CrossFit training on various health markers, such as neuromuscular development, biomechanical/kinematic control, and psychosocial well-being.

Limitations

There are several limitations to our study. The current data were captured in a pediatric and adolescent sports medicine center located in New England. There are no data to support regional different CrossFit training regimens. However, it would be ideal to extract representative samples from various geographical areas. Also, because our clinic focuses on pediatric and adolescent populations, our adult population may not be a truly representative sample. Additionally, severe injury cases associated with CrossFit training that require immediate medical care were not included in this study. Seriously injured CrossFit athletes were most likely transported and treated at urgent care or emergency room. Furthermore, key information such as years of CrossFit training, participation in specialty course, and specific maneuvers at the time of injury was not incorporated in this study. It would be definitely intriguing and likely generate more findings. However, we found that how physicians dictate CrossFit injuries greatly vary; thus, those information was not extracted. Finally, although we made a series of comparisons based on sex and age, this investigation used a retrospective case series study design; thus, this study is more descriptive in nature. To make the current evidence stronger, a prospective study design from representative samples including exposure data are recommended. Also, total number of charts, we included, was relatively small (N = 115). A combination of all of those factors needs to be considered for interpretation of our study.

CONCLUSIONS

This study found that CrossFit athletes aged 19 years and younger have higher odds of sustaining spine-related injuries than those older than 19 years. Additionally, male CrossFit athletes are more prone to shoulder injuries than female CrossFit athletes. Finally, overall lower extremity injuries were greater in females than in males. One study indicated that appropriate supervision of youth is important in helping to reduce injury, particularly in programs with the rigor and intensity of CrossFit.² Thus, future studies need to investigate variables associated with CrossFit-related injuries to aid in the development of programming and prevention of injury.