Generalized joint hypermobility (GJH) has been suggested

Transcription

1 HEIDI SCHMIDT, PT, MHS 1 TRINE LYKKE PEDERSEN, MSS 1 TINA JUNGE, PT, PhD 1,2 RAOUL ENGELBERT, PT, PhD 3,4 BIRGIT JUUL-KRISTENSEN, PT, PhD 1,5 Hypermobility in Adolescent Athletes: Pain, Functional Ability, Quality of Life, and Musculoskeletal Injuries Generalized joint hypermobility (GJH) has been suggested to play a role in the incidence of musculoskeletal injuries in many sports activities. 33 Individuals with GJH have excessive range of motion of joints. 17 Generalized joint hypermobility is typically classified using the Beighton 9-point scoring system, 35 with a score of 4 or greater as the criterion for classification of GJH in adults. 36 The most frequent symptomatic UUSTUDY DESIGN: Cross-sectional. UUBACKGROUND: Generalized joint hypermobility (GJH) may increase pain and likelihood of injuries and also decrease function and health-related quality of life (HRQoL) in elite-level adolescent athletes. UUOBJECTIVE: To assess the prevalence of GJH in elite-level adolescent athletes, and to study the association of GJH with pain, function, HRQoL, and musculoskeletal injuries. UUMETHODS: A total of 132 elite-level adolescent athletes (36 adolescent boys, 96 adolescent girls; mean ± SD age, 14.0 ± 0.9 years), including ballet dancers (n = 22), TeamGym gymnasts (n = 57), and team handball players (n = 53), participated in the study. Generalized joint hypermobility was classified by Beighton score as GJH4 (4/9 or greater), GJH5 (5/9 or greater), and GJH6 (6/9 or greater). Function of the lower extremity, musculoskeletal injuries, and HRQoL were assessed with self-reported questionnaires, and part of physical performance was assessed by 4 postural-sway tests and 2 single-legged hop-for-distance tests. UURESULTS: Overall prevalence rates for GJH4, GJH5, and GJH6 were 27.3%, 15.9%, and 6.8%, respectively, with a higher prevalence of GJH4 in ballet dancers (68.2%) and TeamGym gymnasts (24.6%) than in team handball players (13.2%). There was no significant difference in lower extremity function, injury prevalence and related factors (exacerbation, recurrence, and absence from training), HRQoL, or lengths of hop tests for those with and without GJH. However, the GJH group had significantly larger center-of-pressure path length across sway tests. UUCONCLUSION: For ballet dancers and TeamGym gymnasts, the prevalence of GJH4 was higher than that of team handball players. For ballet dancers, the prevalence of GJH5 and GJH6 was higher than that of team handball players and the general adolescent population. The GJH group demonstrated larger sway in the balance tests, which, in the current cross-sectional study, did not have an association with injuries or HRQoL. However, the risk of having (ankle) injuries due to larger sway for the GJH group must be studied in future longitudinal studies. J Orthop Sports Phys Ther 2017;47(10): doi: /jospt UUKEY WORDS: ballet, Beighton, generalized joint hypermobility, gymnastic, handball, proprioception type of GJH is called hypermobility syndrome (HMS), and for adults, HMS is diagnosed with the Brighton criteria, which are based on a score of 4 or more on the Beighton score and the presence of joint pain in more than 4 joints for a period longer than 3 months. 15 For children, there is no consensus on a specific cut point for GJH, and the criteria for HMS have not yet been defined; therefore, cut points of 5/9, 6/9, and 7/9 on the Beighton score have been suggested. 18 Previous studies have found the prevalence of GJH to range from 2% to 57%, depending on age, sex, ethnicity, and the tests and criteria applied. 36 Within the general population of schoolchildren aged 14 years, GJH (cut point, 4/9) prevalence of 11% to 28% has been reported. 4 Some studies have reported a high prevalence of GJH for certain disciplines, such as dancing and ballet, 7,30 possibly due to the great requirements of flexibility in these activities. 30 However, no study has reported the prevalence of GJH in elite-level adolescent athletes for sports such as team handball, TeamGym gymnastics, as well as ballet. The consequences of GJH are suggested to include joint pain and musculoskeletal injuries, which can result in decreased 1 Department of Sports Science and Clinical Biomechanics, University of Southern Denmark, Odense, Denmark. 2 Health Sciences Research Center, University College Lillebaelt, Odense, Denmark. 3 ACHIEVE Center for Applied Research, Faculty of Health, Amsterdam University of Applied Sciences, Amsterdam, the Netherlands. 4 Department of Rehabilitation, Academic Medical Center, Amsterdam, the Netherlands. 5 Department of Occupational Therapy, Physiotherapy and Radiography, Faculty of Health and Social Sciences, Bergen University College, Bergen, Norway. The Scientific Ethics Committee of the region of Southern Denmark concluded that this project wasn t under the obligation of notification to the Scientific Ethics Committee (number S ). The authors certify that they have no affiliations with or financial involvement in any organization or entity with a direct financial interest in the subject matter or materials discussed in the article. Address correspondence to Heidi Schmidt, University of Southern Denmark, Campusvej 55, Department of Sports Science and Clinical Biomechanics, DK-5230 Odense M, Denmark. heschmidt@health.sdu.dk t Copyright 2017 Journal of Orthopaedic & Sports Physical Therapy 792 october 2017 volume 47 number 10 journal of orthopaedic & sports physical therapy

2 health-related quality of life (HRQoL) 13 and decreased physical activity in both children and adolescents. 11 Despite no increased musculoskeletal pain or decreased perceived physical activity in the general population of schoolchildren aged 8 and 10 years old, 24 GJH was associated with pain development in adolescents aged 14 years. 41 This may be due to imbalance in growth of bone and soft tissues, as well as hormonal changes, resulting in musculoskeletal pain as children with GJH enter puberty. 12 Furthermore, a recent study reported that elite-level adolescent and adult dancers with GJH aged 17 to 27 years have more musculoskeletal complaints, more fatigue, and reduced physical fitness level compared to dancers and nondancers without GJH. 39 Adult elite-level athletes perceive injuries more often than nonelite athletes, and GJH is suggested to lead to an increased injury frequency. 33 Reasons for an increase in injury frequency in those with GJH may include ligamentous laxity and reduced knee joint proprioception, as found previously in a study of children aged 8 to 15 years with HMS. 14 Proprioception is known as an important part of the somatosensory system for controlling balance, with poor balance potentially associated with poor proprioception. 28 Poor balance, measured as increased sway, has been identified in adults with GJH, 31 and a recent study found increased sway to be a predictor of injuries in a general population of children aged 9 to 15 years. 27 However, adult elite-level dancers have shown superior balance ability compared to the general population, possibly because the elite-level training leads to a protective effect against falling, even in dancers with GJH. 5 Whether or not elite-level adolescent athletes with GJH have reduced balance compared to those without GJH is unknown. Additional reasons for sustaining an injury may be decreased muscle strength and consequent decreased muscular stability, as confirmed in some studies of 8-to-10-year-old children with symptomatic GJH 10 and in the general adolescent athlete, 3 but not in studies of children with nonsymptomatic GJH. 22 However, motor performance has never been studied in elite-level adolescent athletes with GJH. The current study aimed to determine the prevalence of GJH in elite-level adolescent ballet dancers, TeamGym gymnasts, and team handball players. A second aim was to investigate whether GJH is associated with self-reported pain, functional ability, HRQoL, injury frequency, postural sway, and motor performance. METHODS Design This is a cross-sectional study of elite-level adolescent athletes with and without GJH, which included ballet dancers, TeamGym gymnasts, and team handball players. Procedures On the day of examination, all adolescents were tested for the presence of GJH by the same physical therapist (H.S.) before completing the motor performance tests. The participants completed 2 electronic questionnaires (HRQoL and injury occurrence) and performed 4 sway tests and 2 motor performance tests. Three physical therapists, 3 MSc students, and 1 TeamGym coach administered the 2 motor performance tests. One associate professor, 1 PhD student, 5 MSc students, 2 physical therapists, and 1 TeamGym coach administered the 2 questionnaires. All examiners and testers were instructed and trained thoroughly in the standardized test procedures prior to the investigation, and examiners, testers, and adolescents were blinded to their mutual test results and GJH status. Groups of 4 to 8 adolescents were tested together, with test durations of 45 to 90 minutes per group. The participants moved through the various testing locations in a semi-random order to maximize time efficiency for testing. The study took place from December 2013 to March 2014 at the participant s usual training facilities. Study Population Participants were elite-level adolescent athletes, aged 13 to 16 years, who participated in ballet, TeamGym gymnastics (noncontact sports), or team handball (contact sport). A total of 132 (36 adolescent boys, 96 adolescent girls) adolescents participated, representing a random selection of 22 of 28 ballet dancers (79%), 57 of 184 TeamGym gymnasts (31%), and 53 of 91 team handball players (58%). The teams were chosen by the researcher and offered participation; both the entire team and the individual athlete within a team could decline to participate. It was, however, not possible to perform analysis of the nonparticipants. Primarily, whole teams participated, with only a few cases of individual nonparticipants. Reasons for not participating were illness, declining to spend the time or not wanting to be absent from training, participation in other studies, or parents declining without reason. The included ballet dancers represented the absolute elite of ballet dancers in Denmark, while TeamGym gymnasts and team handball players represented elite-level teams from two thirds of the country at a minimum level of second division, ranging up to Liga (highest division). When performing sport at this level, a high level of training exposure per week ( hours) was expected and confirmed in all 3 groups. Parents of the participating adolescent gave their written informed consent, and, before testing, each adolescent gave verbal consent to participate, following the Declaration of Helsinki. The Scientific Ethics Committee of the region of Southern Denmark concluded that this project was not under the obligation of notification to the Scientific Ethics Committee (number S ). Clinical Examination The participants were tested to determine their Beighton score, and criteria were applied to determine their status on the Brighton test. 25 A goniometer was used to measure elbow and knee hyperextension above 10, for a potential journal of orthopaedic & sports physical therapy volume 47 number 10 october

3 score of up to 4 points (1 for each joint). Visual observation was used to evaluate the participant s ability to extend the thumb to the volar aspect of the forearm, fifth finger dorsiflexion of greater than 90, and forward bending with touching the palms of the hands to the floor. Participants were classified based on 3 cut points: GJH4 for a Beighton score of 4 or higher, GJH5 for a Beighton score of 5 or higher, and GJH6 for a Beighton score of 6 or higher. Demographics (sex, age, body mass, and height) were obtained. Finally, the dominant leg was identified based on the question, Which leg do you usually kick a ball with? Questionnaires The Rheumatoid and Arthritis Outcome Score for Children (RAOS-Child) Version LK1.0 was used for measuring perceived leg function and HRQoL. 34 The RAOS- Child was developed for children and is presented in the same format as the Knee injury and Osteoarthritis Outcome Score for Children (KOOS-Child). The KOOS- Child has been validated in children, but only covers questions regarding the knee. 32 The RAOS-Child, modified from the KOOS-Child, covers questions regarding the hip, knee, and ankle. Similar modifications have been performed to the KOOS questionnaire for adults 37 to create the RAOS, a valid, reliable, and responsive outcome measurement. 1 These properties have not been tested for the RAOS-Child, but it is assumed that the questionnaire has similar properties to those of the adult version. As in the RAOS, the RAOS-Child consists of 5 domains: pain, other symptoms, function in daily living, function in sport and recreation, and leg-specific quality of life ( with questions on complaints during the most recent week and answers scored from 0 to 4. In total, 48 questions are answered, with the total scores for each dimension ranging from 0 to 100 (0, extreme symptoms; 100, no symptoms). In addition to the RAOS-Child, the participants completed an injury questionnaire with a 1-month recall. The questions included number and location of acute and overuse injuries, recurrence of an acute or overuse injury, exacerbation of an acute or overuse injury (yes/no), and the amount of absence from training and competition due to injuries, rated as not absent, partly absent, and totally absent. The questionnaire was developed and modified from existing studies on injuries among ballet, TeamGym, and team handball participants. 2,29,33 The questionnaire was pretested on a similar group of adolescents with respect to understanding the definitions of injuries. After the pretest, a few words were adjusted. The following definitions were used: acute pain/ injury, a pain/injury occurring suddenly in the context of a known action/exercise ; overuse injury, a pain in the muscles, tendons, or joints that develops over a period without any known reason ; recurrence of injury/pain, pain/damage of the same type and location as experienced previously ; exacerbation of injury/pain, a known injury/pain that is being exacerbated by known or unknown cause. Static Motor Performance The 4 postural-sway tests performed were the 2-legged stance with eyes open (2EO), 2-legged stance with eyes closed (2EC), 1-legged stance on the nondominant leg with eyes open (1EO), and 1-legged stance on the nondominant leg with eyes closed (1EC). Postural sway was assessed with a portable Wii Board (Nintendo, Kyoto, Japan; area, cm), at a sampling frequency of 20 Hz. This device has been shown to have satisfactory reliability and validity in children and adolescents. 26 The Wii Board measures center of pressure (COP), a measure of ground reaction force vector produced by movements of body segments. Postural sway is determined by the excursion of the COP over a fixed period of time. 6 The Wii Board was connected to a laptop computer, with a custom-built program called Sway With Wii calculating the 95% confidence ellipse area and the anteroposterior and mediolateral displacements of the COP. The COP path length (COPL) was further calculated as the total distance traveled by the COP over 1 trial (30 seconds). Each trial was performed for 30 seconds, with bare feet, and with the participant looking at a fixed, eye-level target placed approximately 2.5 m away, with arms crossed at the chest, as described in previous studies. 21 In the 2EO and 2EC conditions, the participants stood with heels and toes together; in the 1EO and 1EC conditions, the contralateral leg s big toe was placed on the medial malleolus of the test leg. One trial of the 2EO condition was performed to familiarize the participant with the testing, while the remaining 3 balance conditions were measured for 3 successful trials, allowing for a maximum of 3 failures (eg, excessive movements of the arms or legs from the original posture, touch-down by the foot on the floor or the measurement equipment, opening of the eyes during 2EC or 1EC tests, or placing more than the big toe on the test leg). If the participant failed to complete the test more than 3 times, the test was canceled and the examiner proceeded to the next test. Between each trial, the participant rested for approximately 30 seconds. Dynamic Motor Performance To assess dynamic motor performance, participants performed 2 hop tests: 1-leg single hop for distance (OLHD) and 1-leg triple crossover hop for distance (COH). Each test was performed 3 times in the above order. Both tests have shown satisfactory reliability. 38 For the OLHD, the participant stood barefoot on the test leg, with his or her toe behind the start line. The participants were instructed to hop forward as far as possible, allowing the use of their arms, land on the same leg, and keep their balance for at least 2 seconds after landing. The length of the OLHD was measured in centimeters from the big toe before takeoff to the heel after landing. For the COH, the participant stood barefoot on the test leg along a 7-m line 794 october 2017 volume 47 number 10 journal of orthopaedic & sports physical therapy

4 on the floor, with the toe behind the start line. When hopping on the right leg, the initial start location was on the right side of the line, and vice versa for the left leg. The participant was instructed to perform 3 consecutive 1-legged maximal hops forward, each time crossing the line on the floor and allowing use of the arms, and to keep his or her balance for at least 2 seconds after landing the third hop. When completing the third consecutive hop, the total length of the COH was measured from the big toe before take-off to the heel after landing. If the participant did not keep his or her balance at landing, or if there was a delay between hops, the trial was disregarded. The participant was allowed to practice the test no more than twice with each leg, and both mean and maximum distances were calculated. Data Analysis Descriptive statistics were summarized using frequency tables and means. For demographic data, group differences by level of GJH status were tested using univariate generalized linear model (GLM) regression (continuous) and logistic regression (nominal), with sport type as the fixed factor. The final GLM, a mixed model, included self-report variables (questionnaires) and measured variables (static sway and hop lengths, one at a time) as TABLE 1 dependent factors; GJH status (GJH4, GJH5, GJH6, one at a time), type of sport, and sex as fixed factors; and age and body mass index as covariates. The identification number was inserted as a random factor and the trial number as a repeated factor in the final GLM. Only 1 trial was performed for the 2EO condition, maximum OLHD, and maximum COH; therefore, a GLM without repeated factors was used for analyses of these variables and the variables from the questionnaires. For injury frequency and location, logistic regression analyses (backward likelihood ratio) were used, with sex and sport as categorical factors and age and body mass index as covariates. Based on the power calculations, at least 26 adolescents with GJH and 79 without GJH were required to show a difference of 29% in injury prevalence between groups, 2 with a power of 80% and a significance level of.05. All analyses were performed in SPSS Version 22 (IBM Corporation, Armonk, NY), and P values less than.05 (2 tailed) were considered statistically significant. RESULTS Prevalence of GJH in elite-level adolescent athletes was 27.3%, 15.9%, and 6.8% for GJH4, GJH5, and GJH6, respectively (TABLE 1). The participants with and without GJH were comparable for all demographic variables, except for an increased prevalence of girls in all GJH groups (TABLE 1). There was a significant difference between sports, with a higher prevalence of GJH4 in ballet (68.2%) and TeamGym (24.6%) compared to team handball (13.2%). The prevalence of GJH5 and GJH6 was significantly higher for ballet than for team handball (GJH5, P =.004; GJH6, P =.024), while sex distribution was similar. There was no significant difference in lower extremity HRQoL between those with and without GJH (TABLE 2), but significant effects of sports type were identified for the RAOS other symptoms subscale and the RAOS leg-specific quality of life subscale, with lower scores for ballet and TeamGym than for team handball. No significant difference between those with and without GJH was found in perceived injuries and related factors (exacerbation, recurrence, or absence from training) between groups, regardless of the cut point (TABLE 3). In all balance tests, athletes with GJH had significantly larger COPL than those without GJH (TABLE 4). For the 1EC test, fewer participants with GJH completed at least 1 of the trials (GJH4: 72% versus 83%, P =.006; GJH5: 57% versus 84%, P<.001). Significant effects were seen for sex and sport for 1EO (TeamGym gymnasts and boys having the greatest static Demographics of Elite-Level Adolescent Athletes Based on 3 GJH Cut Points* Variables GJH4 No GJH4 P Value GJH5 No GJH5 P Value GJH6 No GJH6 P Value GJH status, n (%) 36 (27.3) 96 (72.7) 21 (15.9) 111 (84.1) 9 (6.8) 123 (93.2) Girls, n (%) 33 (91.7) 63 (65.6) (95.2) 76 (68.5).01 9 (100) 87 (70.7).03 Age, y 13.9 ± ± ± ± ± ± Height, cm ± ± ± ± ± ± Weight, kg 51.1 ± ± ± ± ± ± Body mass index, kg/m ± ± ± ± ± ± Training, h/wk 11.6 ± ± ± ± ± ± Abbreviations: GJH, generalized joint hypermobility; GJH4, generalized joint hypermobility with a Beighton score of 4 or higher; GJH5, generalized joint hypermobility with a Beighton score of 5 or higher; GJH6, generalized joint hypermobility with a Beighton score of 6 or higher. *Values are mean ± SD unless otherwise indicated. Significant difference (P<.05) between those with and without GJH. Significant effect for sports type (P<.05). journal of orthopaedic & sports physical therapy volume 47 number 10 october

5 TABLE 2 Perceived Lower Extremity Function Based on the RAOS* RAOS Domain GJH4 (n = 36) No GJH4 (n = 95) P Value GJH5 (n = 21) No GJH5 (n = 110) P Value Abbreviations: GJH, generalized joint hypermobility; GJH4, generalized joint hypermobility with a Beighton score of 4 or higher; GJH5, generalized joint hypermobility with a Beighton score of 5 or higher; GJH6, generalized joint hypermobility with a Beighton score of 6 or higher; RAOS, Rheumatoid and Arthritis Outcome Score. *Values are mean ± SD unless otherwise indicated. Score ranges from 0 to 100, where 0 is worst function and 100 is best function. GJH6 (n = 9) No GJH6 (n = 122) P Value Other symptoms 86.9 ± ± ± ± ± ± Pain 89.1 ± ± ± ± ± ± Function in daily living 97.4 ± ± ± ± ± ± Function in sport and recreation 88.6 ± ± ± ± ± ± Leg-specific quality of life 77.2 ± ± ± ± ± ± TABLE 3 GJH4 (n = 35) Acute and Overuse Injuries for Elite-Level Adolescent Athletes, Based on 3 GJH Cut Points* No GJH4 (n = 92) P Value GJH5 (n = 20) No GJH5 (n = 107) P Value Abbreviations: GJH, generalized joint hypermobility; GJH4, generalized joint hypermobility with a Beighton score of 4 or higher; GJH5, generalized joint hypermobility with a Beighton score of 5 or higher; GJH6, generalized joint hypermobility with a Beighton score of 6 or higher. *Values are n (%) unless otherwise indicated. GJH6 (n = 8) No GJH6 (n = 119) P Value Pain in last year 5 (14.3) 11 (12).99 2 (10) 14 (13.1).45 0 (0) 16 (13.4).11 Acute/overuse injury 20 (57.1) 56 (60.9) (55) 65 (60.7).23 4 (50) 72 (60.5).27 Any injury 22 (62.9) 59 (64.1) (60) 69 (64.5).28 4 (50) 77 (64.7).18 Acute injuries 6 (17.1) 18 (19.6).65 4 (20) 20 (18.7).98 2 (25) 22 (18.5).56 Exacerbation (yes) 2 (5.7) 3 (3.3).99 2 (10) 3 (2.8).35 1 (12.5) 4 (3.4).63 Recurrence (yes) 2 (5.7) 9 (9.8).22 2 (10) 9 (8.4).96 1 (12.5) 10 (8.4).66 Absence (yes) 4 (11.4) 13 (14.1).23 2 (10) 15 (14).55 1 (12.5) 16 (13.4).64 Lower extremity (yes) 3 (8.6) 7 (7.6).83 3 (15) 7 (6.5).24 2 (25) 8 (6.7).12 Upper extremity (yes) 2 (5.7) 7 (7.6).99 2 (10) 7 (6.5).33 1 (12.5) 8 (6.7).19 Back (yes) 2 (5.7) 0 (0).02 1 (5) 1 (0.9).64 0 (0) 2 (1.7).34 Neck (yes) 1 (2.9) 1 (1.1).41 0 (0) 2 (1.9).40 0 (0) 2 (1.7).75 Overuse injuries 17 (48.6) 45 (48.9) (50) 52 (48.6).44 4 (50) 58 (48.7).44 Exacerbation (yes) 3 (8.6) 11 (12).49 3 (15) 11 (10.3).59 1 (12.5) 13 (10.9).76 Recurrence (yes) 10 (28.6) 25 (27.2).54 7 (35) 28 (26.2).53 3 (37.5) 32 (26.9).70 Absence (yes) 3 (8.6) 19 (20.7).58 1 (5) 21 (19.6).35 1 (12.5) 21 (17.6).98 Lower extremity (yes) 17 (48.6) 30 (32.6) (50) 37 (34.6).97 4 (50) 43 (36.1).84 Upper extremity (yes) 3 (8.6) 9 (9.8).36 2 (10) 10 (9.3).56 1 (12.5) 11 (9.2).68 Back (yes) 1 (2.9) 7 (7.6).49 1 (5) 7 (6.5).99 0 (0) 8 (6.7).39 Neck (yes) 1 (2.9) 0 (0) (0) 1 (0.9).02 0 (0) 1 (0.8).34 sway) and 1EC (team handball players and boys having the greatest static sway) conditions. Overall, there was no significant difference between those with and without GJH for the 2 hop tests (OLHD and COH) (TABLE 4). DISCUSSION For ballet dancers and TeamGym gymnasts, the current prevalence of GJH4 was higher than for team handball players. Furthermore, for ballet dancers, the prevalence of GJH5 and GJH6 was higher than for team handball players and the general adolescent population. The injury frequency, HRQoL, and functional muscle performance were similar between those with and without GJH, but those with GJH demonstrated larger sway in the balance tests. 796 october 2017 volume 47 number 10 journal of orthopaedic & sports physical therapy

6 TABLE 4 Postural Sway and Hop Length for Elite-Level Adolescent Athletes, Based on 3 GJH Cut Points* Sway, mm 2EO 2EC 1EO 1EC Hop, cm OLHD COH GJH4 (n = 36) Abbreviations: 1EC, 1-legged with eyes closed; 1EO, 1-legged with eyes open; 2EC, 2-legged with eyes closed; 2EO, 2-legged with eyes open; AP, anteroposterior; COH, 1-legged triple crossover hop; COPL, center-of-pressure path length; GJH, generalized joint hypermobility; GJH4, generalized joint hypermobility with a Beighton score of 4 or higher; GJH5, generalized joint hypermobility with a Beighton score of 5 or higher; GJH6, generalized joint hypermobility with a Beighton score of 6 or higher; L, left; ML, mediolateral; OLHD, 1-legged hop for distance; R, right. *Values are mean ± SD unless otherwise indicated. As previously described, participants in specific elite sports like ballet have a higher prevalence of GJH; however, there is a clear decline in prevalence from the student to the professional level. 30 The high prevalence of GJH4 found in ballet (68.2%) and TeamGym (24.6%) athletes is consistent with previous studies of elite-level adolescent ballet dancers 30 and gymnasts, but higher than the prevalence found in general adolescent athletes. 8 The high prevalence in ballet may partly be due to the inclusion of No GJH4 (n = 95) P Value GJH5 (n = 21) No GJH5 (n = 110) P Value forward flexion in the Beighton score, which is known to correlate highly with being a ballet dancer. 36 Although warmup and training sessions differed between and within sports groups at the time of testing, this did not likely bias the data in relation to prevalence of GJH. Other factors such as sex could not explain the different prevalence rates, 36 because sex distribution of participants was equal across the 3 sports. The large prevalence of GJH in ballet dancers suggests that ballet may be a sport in which those with GJH6 (n = 9) No GJH6 (n = 122) P Value COPL 66.6 ± ± ± ± ± ± AP displacement 3.2 ± ± ± ± ± ± ML displacement 2.9 ± ± ± ± ± ± COPL 98.1 ± ± 18.9 < ± ± 18.8 < ± ± 19.7 <.001 AP displacement 4.0 ± ± 0.9 < ± ± ± ± ML displacement 3.9 ± ± ± ± ± ± COPL ± ± 29.8 < ± ± 29.1 < ± ± AP displacement 4.79 ± ± ± ± ± ± ML displacement 3.38 ± ± ± ± ± ± COPL ± ± 60.7 < ± ± ± ± AP displacement 7.9 ± ± ± ± ± ± ML displacement 5.1 ± ± ± ± ± ± R leg maximum length ± ± ± ± ± ± R leg mean length ± ± ± ± ± ± L leg maximum length ± ± ± ± ± ± L leg mean length ± ± ± ± ± ± R leg maximum length ± ± ± ± ± ± R leg mean length ± ± ± ± ± ± L leg maximum length ± ± ± ± ± ± L leg mean length ± ± ± ± ± ± GJH often choose to participate, and that GJH may help in becoming a successful ballet dancer. TeamGym routines also require flexibility, and the prevalence of GJH4 is remarkably higher in adolescent gymnasts than in the general adolescent population, but not as high as it is in adolescent ballet dancers. An explanation for that may be that GJH is only helpful in the younger gymnasts, for whom the trampoline and tumbling are not so demanding. But, with increasing age and the demands of muscular strength, journal of orthopaedic & sports physical therapy volume 47 number 10 october

7 power, and stability, GJH may no longer prove to be an advantage. Between the groups with and without GJH, no difference was found with respect to perceived leg function, including pain and HRQoL, which contrasts the findings of a recent study showing that GJH in the general population of schoolchildren aged 8 or 10 years may be a contributing and predictive factor in pain development between 13 and 15 years of age. 41 In our participants, the high training volume may be protective against pain and injuries, consistent with a previous study indicating that schoolchildren with GJH and high physical fitness generally had fewer musculoskeletal complaints. 16 Conversely, vigorous exercise may lead to short-term lower-limb pain in schoolchildren. 9 The current findings could also reflect an altered perception of pain and HRQoL in elite-level adolescent athletes, because athletes generally seem to have higher HRQoL than nonathletes. 34 It is noted that in the present study, ballet and TeamGym participants had lower HRQoL than team handball players, as also previously reported. 2,30,39 The current study did not find increased self-reported injury frequency in elite-level adolescent athletes with GJH, which is consistent with the results of a recent study on children. 19 However, this is in contrast to the results of a previous systematic review, in which GJH was associated with an increased risk of knee injuries in athletes from a wide age range (9-39 years old) who participated in different sports, especially contact sports. 33 Our results also contrast with those of a 5-year follow-up study, in which more adult dancers with GJH had multiple joint pain and had to take time off from dancing (greater than 6 weeks) due to injury. 2 The authors of that study hypothesized that injuries in individuals with GJH may take more time to heal, and that there may be greater tissue damage before an injury is reported. 2 Overall, it appears that participation in sports at a high level leads to an increased risk of injuries for adults, which may not be the case for the general child population and for those who participated in our study. In this study, COPL was significantly larger in those with GJH for all balance tests, consistent with recent findings in schoolchildren of the same age with GJH. 23 The increased sway, especially with the eyes closed, may reflect altered proprioception in those with GJH, consistent with results from symptomatic individuals with GJH. 14 Contrary to a previous report that increased postural sway is strongly associated with increased risk of lower-limb injuries, 27 participants in the current study did not report higher incidence of pain or injury. The participants with and without GJH performed similarly on the 2 hop tests, which is in contrast with the results of previous studies of a mixed group of adolescents with GJH who had decreased jumping and walking capacity. 39,40 Despite similar performance, it is possible that muscle activation strategies differed between groups, 20 a line of investigation deserving additional attention. The participation rate was different among sports, with a 79% participation rate for ballet, 31% for TeamGym, and 58% for team handball. A sensitivity analysis showed no difference in factors such as age, sex, and body mass index between those who participated and those who did not. Different sports may reflect different training cultures. A potential limitation may be the use of the current self-reported injury questionnaire, which has not yet been tested for psychometric properties. To minimize the risk of information/interpretation bias, the testers read the injury questionnaire to the participant and were therefore available to assist in interpretation, if necessary. The short-term recall period of 1 month was used to optimize recall of injuries. 1 Data collection occurred outside the period of events/competitions, and therefore may not fully represent the more intense period of activities for elite-level adolescent athletes. The required sample size of 26 participants with GJH was reached when using GJH4 as the criterion, but not for GJH5 and GJH6, potentially limiting the ability to identify differences between groups when using these latter 2 criteria. Strengths of the study include standardized testing procedures, blinding of testers and participants, training of the testers, and the use of reliable and valid objective tests. 41 Validity of the classification of GJH was optimized by using a single examiner. 25 The interpretation of the data is limited to elite-level adolescent athletes participating in sports such as ballet, TeamGym, and team handball. CONCLUSION For ballet dancers and TeamGym gymnasts, the prevalence of GJH4 was higher than for team handball players. Furthermore, in ballet dancers, the prevalence of GJH5 and GJH6 was higher than that in team handball players and the general adolescent population. The injury frequency, HRQoL, and functional muscle performance were not different between those with and without GJH. Those with GJH demonstrated greater sway in the balance tests, which, in the current cross-sectional study, was not associated with injury prevalence and HRQoL. However, the possible risk of injuries due to greater sway in this group must be studied in future longitudinal studies. t KEY POINTS FINDINGS: Elite-level adolescent athletes with and without generalized joint hypermobility participating in ballet, TeamGym, and team handball had no significant difference in incidence of injury, pain, self-rated function, healthrelated quality of life, and hop tests. IMPLICATIONS: The current sports participants with generalized joint hypermobility did not have higher prevalence of adverse health outcomes in this crosssectional study. CAUTION: The study was limited to a 1-month recall period for injury and included individuals participating in 798 october 2017 volume 47 number 10 journal of orthopaedic & sports physical therapy

8 only 3 sports and who were of a homogeneous age group, limiting the external validity of the results. Furthermore, the current injury questionnaire has not yet been tested for psychometric properties. ACKNOWLEDGMENTS: The authors thank the participating ballet dancers, TeamGym gymnasts, and team handball players; physical therapist Charlotte Anker-Petersen from the Royal Danish Ballet School in Copenhagen; and the testers and engineers from the University of Southern Denmark (Odense, Denmark). REFERENCES 1. Bremander AB, Petersson IF, Roos EM. Validation of the Rheumatoid and Arthritis Outcome Score (RAOS) for the lower extremity. Health Qual Life Outcomes. 2003;1:55. org/ / Briggs J, McCormack M, Hakim AJ, Grahame R. Injury and joint hypermobility syndrome in ballet dancers a 5-year follow-up. Rheumatology (Oxford). 2009;48: org/ /rheumatology/kep Brumitt J, Heiderscheit BC, Manske RC, Niemuth PE, Rauh MJ. Lower extremity functional tests and risk of injury in Division III collegiate athletes. Int J Sports Phys Ther. 2013;8: Clinch J, Deere K, Sayers A, et al. Epidemiology of generalized joint laxity (hypermobility) in fourteen-year-old children from the UK: a population-based evaluation. Arthritis Rheum. 2011;63: art Crotts D, Thompson B, Nahom M, Ryan S, Newton RA. Balance abilities of professional dancers on select balance tests. J Orthop Sports Phys Ther. 1996;23: org/ /jospt Danna-Dos-Santos A, Degani AM, Zatsiorsky VM, Latash ML. Is voluntary control of natural postural sway possible? J Mot Behav. 2008;40: JMBR Day H, Koutedakis Y, Wyon MA. Hypermobility and dance: a review. Int J Sports Med. 2011;32: Decoster LC, Vailas JC, Lindsay RH, Williams GR. Prevalence and features of joint hypermobility among adolescent athletes. Arch Pediatr Adolesc Med. 1997;151: archpedi El-Metwally A, Salminen JJ, Auvinen A, Kautiainen H, Mikkelsson M. Lower limb pain in a preadolescent population: prognosis and risk factors for chronicity a prospective 1- and 4-year follow-up study. Pediatrics. 2005;116: Engelbert RH, Bank RA, Sakkers RJ, Helders PJ, Beemer FA, Uiterwaal CS. Pediatric generalized joint hypermobility with and without musculoskeletal complaints: a localized or systemic disorder? Pediatrics. 2003;111:e248-e Engelbert RH, van Bergen M, Henneken T, Helders PJ, Takken T. Exercise tolerance in children and adolescents with musculoskeletal pain in joint hypermobility and joint hypomobility syndrome. Pediatrics. 2006;118:e690-e org/ /peds Falciglia F, Guzzanti V, Di Ciommo V, Poggiaroni A. Physiological knee laxity during pubertal growth. Bull NYU Hosp Jt Dis. 2009;67: Fatoye F, Palmer S, Macmillan F, Rowe P, van der Linden M. Pain intensity and quality of life perception in children with hypermobility syndrome. Rheumatol Int. 2012;32: Fatoye F, Palmer S, Macmillan F, Rowe P, van der Linden M. Proprioception and muscle torque deficits in children with hypermobility syndrome. Rheumatology (Oxford). 2009;48: Grahame R, Bird HA, Child A. The revised (Brighton 1998) criteria for the diagnosis of benign joint hypermobility syndrome (BJHS). J Rheumatol. 2000;27: Gyldenkerne B, Iversen K, Roegind H, Fastrup D, Hall K, Remvig L. Prevalence of general hypermobility in year-old school children and impact of an intervention against injury and pain incidence. Adv Physiother. 2007;9: Hakim A, Grahame R. Joint hypermobility. Best Pract Res Clin Rheumatol. 2003;17: Jansson A, Saartok T, Werner S, Renström P. General joint laxity in 1845 Swedish school children of different ages: ageand gender-specific distributions. Acta Paediatr. 2004;93: org/ /j tb02749.x 19. Junge T, Larsen LR, Juul-Kristensen B, Wedderkopp N. The extent and risk of knee injuries in children aged 9-14 with Generalised Joint Hypermobility and knee joint hypermobility - the CHAMPS-study Denmark. BMC Musculoskelet Disord. 2015;16: org/ /s Junge T, Wedderkopp N, Thorlund JB, Søgaard K, Juul-Kristensen B. Altered knee joint neuromuscular control during landing from a jump in year old children with Generalised Joint Hypermobility. A substudy of the CHAMPSstudy Denmark. J Electromyogr Kinesiol. 2015;25: jelekin Juul-Kristensen B, Clausen B, Ris I, et al. Increased neck muscle activity and impaired balance among females with whiplash-related chronic neck pain: a cross-sectional study. J Rehabil Med. 2013;45: org/ / Juul-Kristensen B, Hansen H, Simonsen EB, et al. Knee function in 10-year-old children and adults with Generalised Joint Hypermobility. Knee. 2012;19: Juul-Kristensen B, Johansen K, Hendriksen P, Melcher P, Sandfeld J, Jensen BR. Girls with generalized joint hypermobility display changed muscle activity and postural sway during static balance tasks. Scand J Rheumatol. 2016;45: Juul-Kristensen B, Kristensen JH, Frausing B, Jensen DV, Røgind H, Remvig L. Motor competence and physical activity in 8-yearold school children with generalized joint hypermobility. Pediatrics. 2009;124: Juul-Kristensen B, Røgind H, Jensen DV, Remvig L. Inter-examiner reproducibility of tests and criteria for generalized joint hypermobility and benign joint hypermobility syndrome. Rheumatology (Oxford). 2007;46: Larsen LR, Jørgensen MG, Junge T, Juul- Kristensen B, Wedderkopp N. Field assessment of balance in 10 to 14 year old children, reproducibility and validity of the Nintendo Wii Board. BMC Pediatr. 2014;14: org/ / Larsen LR, Kristensen PL, Junge T, Møller SF, Juul-Kristensen B, Wedderkopp N. Motor performance as risk factor for lower extremity injuries in children. Med Sci Sports Exerc. 2016;48: MSS Lin YL, Karduna A. Errors in shoulder joint position sense mainly come from the glenohumeral joint. J Appl Biomech. 2017;33: Lund SS, Myklebust G. High injury incidence in TeamGym competition: a prospective cohort study. Scand J Med Sci Sports. 2011;21:e439-e org/ /j x 30. McCormack M, Briggs J, Hakim A, Grahame R. Joint laxity and the benign joint hypermobility syndrome in student and professional ballet dancers. J Rheumatol. 2004;31: Mebes C, Amstutz A, Luder G, et al. Isometric rate of force development, maximum voluntary contraction, and balance in women with and without joint hypermobility. Arthritis Rheum. 2008;59: art Örtqvist M, Iversen MD, Janarv PM, Broström EW, Roos EM. Psychometric properties of the Knee injury and Osteoarthritis Outcome Score for Children (KOOS-Child) in children with knee disorders. Br J Sports Med. 2014;48: journal of orthopaedic & sports physical therapy volume 47 number 10 october

9 33. Pacey V, Nicholson LL, Adams RD, Munn J, Munns CF. Generalized joint hypermobility and risk of lower limb joint injury during sport: a systematic review with meta-analysis. Am J Sports Med. 2010;38: org/ / Parsons JT, Snyder AR. Health-related quality of life as a primary clinical outcome in sport rehabilitation. J Sport Rehabil. 2011;20: Remvig L, Jensen DV, Ward RC. Are diagnostic criteria for general joint hypermobility and benign joint hypermobility syndrome based on reproducible and valid tests? A review of the literature. J Rheumatol. 2007;34: Remvig L, Jensen DV, Ward RC. Epidemiology of general joint hypermobility and basis for the proposed criteria for benign joint hypermobility syndrome: review of the literature. J Rheumatol. 2007;34: Roos EM, Roos HP, Lohmander LS, Ekdahl C, Beynnon BD. Knee Injury and Osteoarthritis Outcome Score (KOOS) development of a self-administered outcome measure. J Orthop Sports Phys Ther. 1998;28: org/ /jospt Ross MD, Langford B, Whelan PJ. Test-retest reliability of 4 single-leg horizontal hop tests. J Strength Cond Res. 2002;16: Scheper MC, de Vries JE, de Vos R, Verbunt J, Nollet F, Engelbert RH. Generalized joint hypermobility in professional dancers: a sign of talent or vulnerability? Rheumatology (Oxford). 2013;52: CHECK Your References With the JOSPT Reference Library JOSPT has created an EndNote reference library for authors to use in conjunction with PubMed/Medline when assembling their manuscript references. This addition to Author and Reviewer Tools on the JOSPT website under offers a compilation of all article reference sections published in the Journal from 2006 to date as well as complete references for all articles published by JOSPT since 1979 a total of more than 20,000 unique references. Each reference has been checked for accuracy. This resource is updated quarterly on JOSPT s website. The JOSPT Reference Library can be found at: authors/author_reviewer_tools rheumatology/kes Scheper MC, de Vries JE, Juul-Kristensen B, Nollet F, Engelbert RH. The functional consequences of Generalized Joint Hypermobility: a cross-sectional study. BMC Musculoskelet Disord. 2014;15: org/ / Sohrbeck-Nøhr O, Kristensen JH, Boyle E, Remvig L, Juul-Kristensen B. Generalized joint hypermobility in childhood is a possible risk for the development of joint pain in adolescence: a cohort study. BMC Pediatr. 2014;14: doi.org/ /s MORE INFORMATION october 2017 volume 47 number 10 journal of orthopaedic & sports physical therapy