Injuries in CrossFit practioner: a cross-sectional study

Abstract Aim: The present study aimed to investigate epidemiological parameters associated with the onset of injuries in CrossFit practitioners. Methods: Cross-sectional study, with fifty-two subjects (28 ± 7 years,70 ±13 kg), regular practitioners of CrossFit, of intermediate level. The Rombaldi questionnaire was applied, related to the occurrence of injuries during physical activity. The Shapiro-Wilk normality test, Pearson's chi-square, and Fisher's exact test were used, using p < 0.05. Results: Our findings show that the prevalence of injury in cross-fitters is 38%, having an incidence rate of 3.7 per 1000 h of training. The most recurrent injury was stretching (41%). The most affected regions were the shoulder and lumbar (34%). The exercise model with the highest association with injury development was Olympic weightlifting (p = 0.004). Conclusion: The CrossFit practice showed a moderate prevalence of injuries, stretching the main type. The most affected regions were the shoulders and the lumbar. In relation to the most dangerous exercise model for the appearance of injuries, the Olympic weightlifting exercises stand out.


Introduction
CrossFit is one of the fastest growing training methods for physical conditioning in the world, with great popularity and adherence 1 .Its creation dates from 1995 when it was created by former gymnast Greg Glassman 2 .Such a method is characterized as High-Intensity Functional Training (HIFT) and aims at developing general physical conditioning using increases in physical components such as aerobic fitness, motor coordination, agility, balance, speed, strength, and muscle resistance 3 .Based on this, it is suggested that CrossFit promotes broad and global physical conditioning, preparing practitioners for any physical demands of daily life 2 .Crossfit characteristically employs constant variations between gymnastic movements, Olympic Weightlifting (OWL), and cyclical exercises for aerobic capacities, alternated amongst themselves.A Crossfit workout is typically carried out in 60 min classes, subdivided into the warm-up, technique or strength development, and the main part which is properly destined for high-intensity exercises lasting between 5-20 min 3 .The knowledge of the occurrence of musculoskeletal injuries in modalities that become more popular every day, like CrossFit, is of great importance so that possible preventive measures can be taken with greater effectiveness 5 .Paluska 6 states that any practitioner of regular physical activity is subject to musculoskeletal injuries associated with the practice of the sport or modality.These injuries are potentiated by factors such as poor execution of exercises, improper load progression, and imbalances in training planning.However, although the American College of Sports Medicine recognizes the several beneficial effects of programs of extreme conditioning such as CrossFit, it still stands out as a concern concerning the rate of injuries in practitioners of this modality.Thus, the study aimed to quantify epidemiological parameters of prevalence, rate, CrossFit practitioner injuries incidence, most affected body regions, and exercise models associated with the appearance of injuries in CrossFit practitioners.

Study design
This research is characterized as a quantitative cross-sectional study.Such procedures were performed in accordance with the law 11.794/08resolution 196/96 with the use of human beings and approved by the local Ethics Committee (00918918.3.0000.0119).Prior to the beginning of the study, all participants were informed about the purpose of the study and signed the Informed Consent Form (ICF).
As per the flow chart (Figure 1), 150 questionnaires were delivered in three 97 CrossFit boxes in the local city.Then after the predetermined period for completion and delivery of the questionnaires by the participants, 62 completed questionnaires were returned.After evaluation, 10 questionnaires were excluded for being filled out incorrectly.Thus, 52 CrossFit practitioners from Criciúma -SC of both genders participated in the study.
The participants were selected by the non-probabilistic convenience method 7 , according to the following inclusion criteria: a) being an active practitioner of the sport b) submitting the questionnaire within the established period; d) having at least three months of practice.
And exclusion criteria: a) present some type of chronic nontransmissible disease chronic injury or be a carrier of some physical limitation/disease that interfere in its practice b) have visual or cognitive alterations that may interfere in the understanding of the questionnaire.

Instruments
To measure the prevalence, incidence, type, and body segment most injured during the practice of Cross-Fit, a modified Rombaldi 8 questionnaire was used.The questionnaire was composed of 16 questions, 4 open and 12 closed, described as follows: 1) Anthropometric and sociodemographic characterization; 2) Time practicing of  20 , which uses the triple injury criterion covering a wide variety of injuries that can occur in CrossFit, being described as 1) Total removal from CrossFit training and other activities non-routine physical activities for > 1 week; 2) Modification of normal training activities in duration, intensity or mode for >2 weeks; 3) Any physical complaint serious enough to warrant a visit to a healthcare professional.

Crossfit injury rate
The injury rates were calculated by estimating the training volume of practitioners in the retroactive period of practice of the modality, obtained in the inventory.The total weekly volume of training was multiplied by the number of weeks corresponding to the period of practice.The rate was then converted into the number of injuries/ 1000 h of training 9 .

Statistical analysis
For the statistical analysis, a spreadsheet was assembled for the elaboration of the database, based on the data collection performed.Quantitative variables were presented as mean ± standard deviation and qualitative variables as frequency and percentage.For all statistical analyses, a value of p < 0.05 was adopted to indicate a statistically significant difference or association between groups.Quantitative variables were assessed for normality by applying the Shapiro-Wilk test.Verification of the existence of an association between qualitative variables was performed by applying Pearson's chi-square test and Fisher's exact test, followed by residue analysis when statistical significance was observed.The analyses were performed in the statistical package SPSS (Statistical Package for the Social Sciences) version 20.0.

Characterization of the sample
As shown in Table 1, 52 participants took part in the study, approximately 57% (30 participants) were male and 43% ( 22) female, with a mean age of 28 ± 7 years, a weight of 70 ± 13 kg, and a mean practice time for 14 months.Thirty-two practitioners, representing 64% of the sample, train for over 12 months, indicating intermediatelevel subjects.Most of the practitioners (65%; n = 34) present a high frequency (5 times/week).

Prevalence and injury rates of CrossFit practitioners
The prevalence of injuries was 38% (n = 20) of the total sample (n = 52).Regarding injury rate per 1,000 hours of training, a rate of 3.7/1000 h of training was observed (Table 2).

Exercise models associated with the appearance of injuries
Of the CrossFit practitioners (Table 3) who injured their shoulder, 55% stated that they were injured while performing exercise models that involved Olympic weightlifting exercises, 25% cyclical exercise models, and 45% gymnastic exercise models.In relation to knee injuries, 71% of the practitioners were informed to have been injured during the practice of exercise models involving Olympic weightlifting, 29% during cyclic exercise models, and 17% during gymnastics.Concerning the lumbar lesions, 66% reported having suffered this lesion during the realization of models of exercises involving Olympic weightlifting, 16% in cyclic exercises, and 16% in gymnastics.

Discussion
This study investigated specifically, the prevalence, injury rate, incidence, anatomical regions, and exercise models associated with the development of injurie in CrossFit practitioners.Our findings show a prevalence of 38%, an injury rate of 3.7/1000 h of training, and the highest incidence (41%) of injuries, by stretching.Anatomically the most affected regions were the shoulders and lumbar.In relation to the exercise models that more cause injuries we point out the OWL.All of them suffered some type of injury (muscular/osteoarticular) during the period of CrossFit practice.These data are similar to values that are close to the earliest studies 1,10,11 , which found values between 31 and 37%.However, the injury rate among CrossFit practitioners in this study was 3.7/1000 h of training.According to Ory 12 , every form of physical activity offers some risk of injury to the practitioner.Our injury rate values presented here (3.7/1000 h) are higher than those found in a recent systematic review 13 , which varied from 1.94 to 3.1/1000 h of training.Values presented in n and (%), in the quantitative variables.However, the values presented in the present study are lower when compared with traditional sports such as football, which presents values varying from 5.2/1000 h to 7.8/1000 h of training, and field tennis with values of 4.7/1000 h of training, and similarly found in recreational running 3.6/1000 h 14,15 .
According to Rahusen 18 , these injuries may be related to the excessive stretching of muscle fibers that during the strength exercises used in CrossFit, the total and rapid amplitude is required, in addition to movement speed, causing the practitioner to exceed the maximum amplitude of the movement, leading to musculoskeletal injury, which may lead to a compromise of muscle and joint functionality.
As for the anatomical regions with greater involvement in injuries, our results point to a predominance of the shoulder and lumbar (34%), followed by the knee (23%).Montalvo 9 , when investigating the most injured anatomical region of CrossFit practitioners, concluded that the regions with the highest incidence of injuries are the shoulder and the lumbar spine, as well as Xavier and Costa 3 , who found percentages of 44% and 40%, respectively.Such evidence regarding the location of the lesion corroborates our findings.One explanation for the appearance of injuries in these locations is possibly related to the fact that the practitioners perform the movements repeatedly until concentric fatigue.Specifically in lesions of the shoulder region, when limb movements are performed in abduction/adduction and/or flexion/extension above 90°, these movements are biomechanically considered triggers for lesions due to the disproportionate adjacent angles and forces 19 .
Regarding the exercises that have a greater association with the appearance of injuries in the knee follow-up, our study found a percentage of 71% for Olympic weightlifting and 29% for cyclic exercises.Weisenthal 22 point out the knee as one of the most frequently injured sites during exercise.Montalvo 9 states that Olympic weightlifting movements can facilitate the appearance of injuries when compared to other simpler movements.It is a fact that according to Tavares and oliveira 23 such movements (Olympic lifts) generate a higher risk due to the presence of a longitudinal external load.Another possible justification is linked to intrinsic anatomic factors such as discrepancy of muscular balance and biomechanics that relate to subtalar pronation or supination 22 .

Conclusions
We conclude that the prevalence of injuries in CrossFit practitioners in this study is moderate when compared with other CrossFit studies.Nevertheless, it is lower when compared with traditional sports such as football and tennis.In relation to the rate of injuries, it is above average.As a practical application, we suggest redoubled care in the Olympic weightlifting exercises, having in mind that it appears in first place in the three evaluated segments.

Figure 2 -
Figure 2 -Incidence of injuries in crossfit practitioners.

Figure 3 -
Figure 3 -Anatomical regions affected by the injuries reported by the study participants.

Table 1 -
Characterization of the sample and profile of practitioners.

Table 2 -
Prevalence and injury rate of crossfit practitioners.

Table 3 -
Exercise models are associated with the appearance of injuries.