TITLE PAGE

The links between Generalized Joint Laxity and the incidence, prevalence and severity

of limb injuries related to physical exercise: A systematic literature review

Authors: Alexander Tinglea, Oliver Bennetta, Amy Wallisa, Shea Palmera*

Affiliations: aFaculty of Health and Applied Sciences, University of the West of England,

Bristol, UK

Corresponding Author Details:

*Professor Shea Palmer, Professor of Musculoskeletal Rehabilitation, Department of Allied

Health Professions, University of the West of England, Blackberry Hill, Bristol, BS16 1DD

Tel: +44 117 3288919

Email: shea.palmer@uwe.ac.uk

ORCID: https://orcid.org/0000-0002-5190-3264

Biographical Notes:

Alexander Tingle is a sports rehabilitator currently working as a neurotherapy assistant

within the acute stroke unit at the John Radcliffe hospital, Oxford. His main interests

regarding practice and research are neurorehabilitation and functional biomechanics.

Oliver Bennett graduated from the University of the West of England with a degree in

Sports Therapy and Rehabilitation. He currently works in a duo alongside the Head of

Medical at Cheltenham Town Football Club. As well as rehabilitation his interests also lie

with Sports Biomechanics, hoping to develop that interest through further education to aid his

professional practice.

1

Amy Wallis graduated from the University of the West of England with a First Class

Honours in Sport Therapy and Rehabilitation. She has a keen interest in musculoskeletal

rehabilitation and biomechanics. She is hoping to continue her professional development

through further education in the near future.

Shea Palmer is a physiotherapist currently employed at the University of the West of

England Bristol as Professor of Musculoskeletal Rehabilitation. His main research interests

are related to the assessment and management of musculoskeletal disorders, with a particular

interest in joint hypermobility.

Acknowledgements: The authors would like to acknowledge the support of Sam Smith and

Sarah Fountain in the design and conduct of the review.

Word Count: 3600

2

ABSTRACT

The links between Generalized Joint Laxity and the incidence, prevalence and severity

of limb injuries related to physical exercise: A systematic literature review

Background: Generalized Joint Laxity (GJL) is a significant risk factor for lower limb injury

incidence and prevalence in sporting populations. However, the links with upper limb injury

and injury severity in a wider population of people undertaking physical exercise have not

been systematically reviewed to date.

Objectives: The primary aim was to determine the links between GJL and the incidence,

prevalence and severity of upper and lower limb injuries related to physical exercise.

Secondary aims were to identify the quality of the existing research evidence and gaps within

the literature that may warrant future research.

Methods: Relevant literature was identified using online databases (SportDiscus, Medline,

CINAHL and EMBASE OVID) and snowballing. Research papers with a primary aim of

identifying a link between GJL and upper and lower limb injury incidence, prevalence and/or

severity were included. The population of interest was those undertaking physical exercise,

not limited to sport. Included papers were critically appraised and a narrative synthesis

conducted.

Results: 274 studies were identified. Following application of the inclusion and exclusion

criteria, nine papers were selected for critical appraisal that investigated the link between GJL

and the incidence, prevalence and/or severity of limb injuries. The link between GJL and

lower limb injury incidence and prevalence was supported across numerous types of physical

exercise. However, the links between GJL and upper limb injury, and injury severity was

inconclusive.

3

Conclusions: GJL is associated with increased lower limb injury incidence and prevalence.

Future research should investigate the links between GJL, upper limb injury and injury

severity.

Keywords: Generalized Joint Laxity; Incidence; Prevalence; Injuries; Exercise

Funding Details: No funding was received to support this work.

4

MANUSCRIPT

Introduction

Generalized Joint Laxity (GJL) has been defined as an increased range of motion in multiple

joints relative to the normal population1, although an internationally agreed definition has yet

to be established2. Thus variations of the term have emerged such as Generalized Joint

Hypermobility and Generalized Ligamentous Laxity. For the purposes of clarity and

consistency, the term GJL shall be used throughout this paper.

GJL is recognised as a genetically inherited trait that alters the composition and

alignment of the collagen matrix within connective tissues such as ligaments, ultimately

increasing the range of a joint via enhanced soft-tissue extensibility3. GJL has been reported

to be more common in women and in some ethnic groups such as native American and

African populations4. The prevalence of GJL is suspected to be approximately 10 to 20

percent5 of the general population and it affects both sporting and non-sporting individuals.

To identify GJL, the Beighton score6 is regarded as the most valuable examination

tool in terms of validity and reproducibility, and thus is used most frequently within research

and clinical practice7. The Beighton score assesses the mobility of nine joints, including

bilateral thumb opposition to the forearm, bilateral fifth finger extension, bilateral knee

extension, bilateral elbow extension and lumbar spine/hip flexion in standing. Each

movement can be measured efficiently and reliably within a clinical environment, requiring

only a goniometer to do so8. A score of ≥4/9 is commonly used to identify GJL9.

When identifying GJL it is imperative to distinguish it from other connective tissue

disorders such as Ehlers-Danlos Syndrome, osteogenesis imperfecta, Marfan syndrome,

achondroplasia, and GJL’s symptomatic equivalent, previously known as Joint

Hypermobility Syndrome (JHS)10. The primary distinction between GJL and JHS is that

5

people with the latter usually experience arthralgia, which may result in decreased physical

activity11. Individuals with GJL do not report arthralgia resulting from joint laxity12, except

following acute injury. GJL accompanied by arthralgia for longer than 3 months in four or

more joints were historically the major criteria for the diagnosis of JHS13, although it should

be noted that the diagnostic criteria for syndromic joint hypermobility have recently been

revised14. The present review relates to otherwise healthy people with GJL. It excludes those

with arthralgia associated with recognised connective tissue disorders (such as JHS or Ehlers-

Danlos Syndrome) that have GJL as a distinctive feature.

GJL may be considered advantageous in certain sports that require optimal flexibility

for aesthetic and performance-enhancing purposes, such as gymnastics15. However, GJL may

be detrimental in sports like rugby union, where the link between GJL and musculoskeletal

injury has previously been established16.

Several studies have examined the association between GJL and sporting injury. For

example, Östenberg and Roos17 performed a prospective study on 123 European female

footballers over the course of one season and found that GJL was a significant risk factor for

injury, especially within the knee joint. Myer et al.18 also demonstrated a significantly

increased risk of anterior cruciate ligament (ACL) injury in female football and basketball

players with GJL. In contrast, other studies have disputed the link between GJL and injury,

such as Beynnon et al.19 who demonstrated that GJL had no significant influence on ankle

ligament injury in 118 collegiate soccer, lacrosse and field hockey athletes.

Many existing studies are limited by their focus on conventional sports, excluding

populations with GJL who partake in other forms of physical exercise such as dance20. To

address this, Pacey et al.21 conducted a systematic literature review with meta-analysis of 18

studies, incorporating sporting and physically active populations that ranged from soccer

players22 to ballet dancers23 and military personnel24. The results demonstrated a statistically

6

significant risk of knee joint injury for participants with GJL compared with their non-

hypermobile counterparts (p<0.001). No significant risk was established for ankle joint

injury. Although Pacey et al.21 identified the potential implications of GJL across a range of

physical activities, the review focused on lower limb injury and may now be partially

outdated by the emergence of other research25. For example the links with upper limb injury

have been investigated in a more recent primary study26.

Another consideration is that the link between GJL and severity of injury (such as

sporting and/or working time lost, or permanent damage27) has not previously been

systematically reviewed. Instead, previous research has focused primarily on injury incidence

and prevalence. Prevalence is the proportion of people with an injury at a specified point in

time28, whereas incidence is the number of new injuries occurring across a certain time period

(for example per 1000 hours)27.

The present review focuses on ‘physical exercise’ to include both sporting and

physically active populations who do not partake in a defined sport. The term ‘physical

activity’ refers simply to the movement of skeletal muscles which results in energy

expenditure29. However, physical exercise is a sub-category of physical activity that involves

planned, structured activity, specifically intended to enhance or maintain physical fitness30.

Therefore, the focus on physical exercise increased the specificity of the review by excluding

generic physical activity, whilst simultaneously broadening the scope from conventional

sports.

The primary aim of this review was therefore to establish the links between GJL and

the incidence, prevalence and severity of upper and lower limb injuries related to physical

exercise. Articulated in PICOS format, the research question relates to people with GJL

(‘Population’); physical exercise (‘Intervention’ or exposure); people without GJL

(‘Comparator’); the incidence, prevalence and severity of upper and lower limb injuries

7

(‘Outcomes’); and relevant primary research study designs, such as cohort, case-control or

cross-sectional studies (‘Study’ type). Secondary aims were to establish the quality of the

existing evidence and to identify any gaps that may warrant future research.

Methodology

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA)

statement31 was used to ensure that methodological rigour was maintained and appropriate

reporting of the review32. The protocol was not prospectively registered.

Search Strategy

The search strategy employed four online databases, followed by primary snowballing,

ensuring identification of relevant studies33. The online databases were SportDiscus, Medline

and CINAHL (via the EBSCO search engine) and EMBASE (via OVID). Searches were

concluded in April 2017.

Key search terms were used to identify relevant papers (Table 1). Terms were

stratified into three searches that reflected different aspects of the main question. Search one

represented the population (those with GJL), whilst two and three represented the exposure

(physical exercise) and outcomes (limb injury and Beighton score) respectively34. Papers

associated with recognised connective tissue disorders (such as JHS or Ehlers-Danlos

Syndrome) were excluded. The Boolean operators ‘OR’ and ‘AND’ were used to connect

terms within each search and between searches respectively35. Truncations were used to

account for possible variation in spelling or plurality.

Two pairs of researchers conducted the electronic searches independently before

meeting and comparing results. Minor discrepancies were resolved through discussion and

consensus, and the search findings were then finalised and duplicates removed. This process

8

assisted in reducing potential errors, and increased methodological reproducibility and

rigour36.

Study Selection

Study selection was again performed independently by two pairs of researchers before

meeting, discussing discrepancies and reaching agreement. Predetermined inclusion and

exclusion criteria (Table 2) were applied in turn to the titles, abstracts and full texts. Primary

snowballing was then used to identify further potential papers from the reference lists of

remaining studies to maximise the probability that all relevant literature had been identified37.

The same protocol for applying the inclusion and exclusion criteria was repeated on those

studies.

Critical Appraisal Tool

Scottish Intercollegiate Guidelines Network (SIGN) checklists

(http://www.sign.ac.uk/index.html) were selected as critical appraisal tools, facilitating an

objective, structured and standardised review of the studies38. This included judgements

related to potential bias. As before, two pairs of researchers independently critiqued each

paper before meeting to resolve disagreements by consensus. Key information about each

study was extracted and presented as a narrative synthesis. Extracted information was

tabulated to facilitate comparison across studies. Information included the study design,

country in which the research was conducted, the stated aims, which limbs were investigated,

the Beighton cut-off score used, sample size, sex and age of participants, type of physical

exercise, outcome categories (i.e. incidence, prevalence or severity) and the main statistical

findings. Meta-analysis was not possible due to wide heterogeneity in methods and outcomes.

9

Results

Figure 1 illustrates the selection procedure. The online database searches retrieved 396

potential studies, with a further 25 potential papers identified via primary snowballing. 274

papers remained following exclusion of duplicates. Following application of the inclusion

and exclusion criteria, nine studies were included in the review (Akodu et al.39; Bin Abd

Razak et al.40; Blokland et al.41; Decoster et al.42; Konopinski et al.43; Saremi et al.26; Smith et

al.44; Stewart and Burden16 and Sueyoshi et al.45).

Data Extraction

Key information from the included studies is displayed in Table 3. Studies were conducted in

many countries across the world and included military recruits and participants from a wide

range of sports. Evidence related to injury prevalence predominated, as did evidence related

to the lower limb. Total sample sizes ranged from 47 to 310. Generally, evidence supported

links between GJL and lower limb injury incidence and prevalence. Evidence related to

injury severity or related to links between GJL and upper limb injury was inconclusive.

Critical Appraisal

The five cohort studies16,39,41,42,43 and two cross-sectional studies26,44 were assessed using the

cohort study SIGN checklist (Table 4a). The two case-control studies40,45 were appraised

using the case-control checklist (Table 4b). A general indicator of quality (risk of bias or

confounding) for each study is provided by criterion 2.1 in Tables 4a and 4b. On this basis,

the studies by Blokland et al.41, Konopinski et al.43 and Bin Abd Razak et al.40 were identified

as the highest quality papers. Common shortcomings across studies related to blinding and

accounting for potential confounders in the design and analysis.

10

Discussion

The primary aim of this systematic review was to examine the link between GJL and the

incidence, prevalence and severity of limb injuries related to physical exercise. Nine studies

were reviewed. Due to the multiple outcome measures observed, there was a degree of

disparity between the studies’ results. Generally, there was broad evidence to support the

links between GJL and the prevalence 26,39,40,43,44,45, incidence16,42,43 and severity26,43 of limb

injuries. However, Blokland et al.41 failed to demonstrate a statistically significant link

between GJL and their outcome categories related to injury incidence and severity. Decoster

et al.42 only supported a link between GJL and ankle injury incidence, but not overall injury

incidence or severity.

Further disparity was evident when the nature and location of injuries were analysed.

Overall, evidence related to lower limb injuries predominated in the literature. For example

Akodu et al.39 showed a significant relationship between GJL and lower limb injury

prevalence but not upper limb injury prevalence. Saremi et al.26 displayed significant

differences between GJL and non-GJL participants for chronic, but not acute, shoulder injury

prevalence. In some studies it was not possible to isolate the links between GJL and upper or

lower limb injury due to pooling of injury data. Additionally, only four of the studies

investigated injury severity, two of which supported a link with GJL26,43 and two of which

disputed a link41,42. Thus, evidence related to upper limb injury and injury severity was

relatively inconclusive.

Study Quality

Three types of observational studies were reviewed. Five were cohort studies39,41,42,16,43, two

were case-control studies40,45, and two were cross-sectional studies26,44. The cohort study

design is recognised for producing the highest level of evidence among observational

11

studies46. Thus, this ratio of study designs was considered beneficial for the overall quality of

the systematic review.

One strategy to achieve high external validity is recruiting a large sample size, as that

increases the likelihood that participants are representative of the target population, and

provides allowance for drop-outs47. On this basis, Decoster et al.42 arguably had high external

validity as they observed 310 participants, succeeded by Smith et al.44 and Bin Abd Razak et

al.40 who recruited 200 participants each. The remaining sample sizes ranged from 118 to 47

participants. However, small samples may still possess sufficient numbers to produce

statistically significant outcomes and the required sample may be determined via sample size

calculations48. Recruiting to sample size calculations enables researchers to formulate robust

conclusions from the data and permits the generalisation of results49. Only Saremi et al.26 and

Bin Abd Razak et al.40 conducted sample size calculations to inform recruitment of an

appropriate number of participants.

The age of participants and type of physical exercise investigated represented other

important considerations in terms of external validity. For example, the mean age of

participants in Decoster et al.42 was 20 years and only lacrosse players were observed.

However, it is important to acknowledge that their aim was specific to lacrosse players with

GJL, and they sampled from 17 different teams. Therefore, although the results may not be

applicable to other sports, they may still be generalisable in the context of their target

population, enhancing their external validity47. The majority of the other studies also included

only one type of physical exercise and relatively small age ranges, focusing predominantly on

adolescents and young adults. One exception was Saremi et al.26, who, alongside a mixed-

gender sample, incorporated a total of 14 types of physical exercise and an age range of 17 to

37 years. On initial review, this appears to substantially increase this study’s external validity,

however the type of physical exercise differed vastly, ranging from 24 martial artists to only

12

2 boating athletes, and the number of male subjects (n=80) was disproportionate to females

(n=38). Thus, while the sample was diverse, the external validity of these results should be

interpreted with caution50.

A primary threat to internal validity are confounding variables52, factors other than

GJL that disproportionately affect the incidence, prevalence or severity of injury between

participant groups. For example, Blokland et al.41 identified age, body mass index and soccer

exposure as potential confounders for injury incidence measures, and they accounted for

these in their regression analysis. Smith et al.44 also employed regression analysis to account

for confounding. However, in the remaining seven papers only two acknowledged possible

confounding factors45,16 and none accounted for these in analysis, reducing internal validity53.

Implementation of inclusion and exclusion criteria can also control for potential

confounding variables51. Saremi et al.26 excluded participants if they had any deformity or

disorder that interfered with the Beighton score assessment. Bin Abd Razak et al.40 was the

only other primary study to specify inclusion and exclusion criteria aimed at controlling

potential confounders.

Blinding is another strategy for enhancing internal validity, where deemed feasible51.

Four studies40,41,43,45 utilised blinding of the researchers who recorded data (either to

hypermobility status or injury history). However, none of the studies blinded the participants

to study aims. This renders the studies which retrospectively recorded injuries via

questionnaires and interviews39,40,44,45 susceptible to obsequiousness bias, in which participants

may have falsely provided previous injury information during data collection to appear

helpful to the study aims54. Additionally, unrelated to blinding, retrospective studies may

have been vulnerable to recall bias, as participants could have forgotten previous injuries,

further affecting internal validity55.

13

Despite this, the majority of studies standardised their methods of injury recording,

increasing inter-rater and intra-rater reliability56. However, Sueyoshi et al.45 asked participants

to detail their own medical history, whilst Stewart and Burden16 and Bin Abd Razak et al.40

both provided definitions of what was considered to be an injury but provided few other

details on how injuries were identified and recorded. Consequently, it is difficult to assess

how robust the procedures were in comparison to those who implemented standardised injury

questionnaires or forms57.

Heterogeneity

A number of factors contributed towards heterogeneity of the studies, making comparison

and interpretation of the overall findings challenging51.

Outcome Categories

The review incorporated three categories of outcomes, namely incidence, prevalence and

severity of injury. However, a range of different measurement tools was used, impacting on

the comparability between studies. The exceptions to this were Konopinski et al.43, Akodu et

al.39 and Blokland et al.41 who all registered injuries via the Fédération Internationale de

Football Association Medical Assessment and Research Centre (F-MARC) consensus

statement injury form58. However, Konopinski et al.43 and Blokland et al.41 measured injury

incidence and severity using the F-MARC form, whilst Akodu, et al.39 used it to record injury

prevalence. Thus the results cannot be directly compared between studies.

In the studies that measured injury prevalence, discrepancies were identified in the

length of time that participants were observed. Bin Abd Razak et al.40 and Saremi, et al.26 both

measured prevalence over a six-month period, whereas Akodu, et al.39 conducted a 12-month

investigation, Smith et al.44 measured over two weeks, and Sueyoshi, Emoto and Yuasa45

14

failed to clearly state their time frame. However, it is important to note that these studies still

produced statistically significant results supporting the link between GJL and injury

prevalence, regardless of the duration of observation, arguably reducing the impact of the

different time frames on the comparability of results59.

Conversely, the studies which observed injury incidence16,41,42,43 all measured the

occurrence of new injuries at a standard rate of per 1000 hours. , tTherefore the results of

these studies may be more readily compared, with three studies supporting a link between

GJL and injury incidence16,42,43 and one finding no link41.

There was also a range of different methods used to measure injury severity. As

discussed previously, Konopinski et al.43 and Blokland et al.41 both used the F-MARC form to

measure their outcomes, where severity was recorded as: Slight (0 days), minimal (1-3 days),

mild (4-7 days), moderate (8-28 days) and severe (>28 days). Decoster et al.42 also measured

time loss but using different categories. Saremi et al.26 assessed severity in terms of function,

using the QuickDASH questionnaire (Disability of the Arm Shoulder and Hand) and shoulder

instability. The use of different measures negatively impacts on comparability across studies.

Participant Age

Regarding participant age, the greatest differences can be observed between Smith et al.44,

who recruited participants aged six to 16 years, and Akodu et al.39, who recruited participants

aged 18 to 38 years. Whilst the difference in ages between these studies clearly reduces their

comparability from a sampling perspective, evidence also suggests that joint laxity decreases

with age, due to reductions in soft-tissue extensibility caused by increased collagen fibre

diameter and elastin fibre degeneration60. Therefore, the extent of GJL in older participants

may have differed to that of younger participants, potentially also impacting on the links with

15

limb injury. Unfortunately, this potential issue could not be explored further in this review

due to lack of relevant details reported within the studies.

Nature of physical exercise

The types of physical exercise investigated included individual studies related to cricket39,

lacrosse42, military training40, netball44, rugby16, soccer41,43 and volleyball45. One study26

included participants participating in a wide range of activities (basketball, boating, boxing,

climbing, fitness, football, handball, martial arts, mountaineering, racket sports, swimming,

volleyball, water-polo and wrestling). The nature of the physical exercise undertaken may

have affected the comparability of the studies. For example, high-contact sports, such as

rugby union and soccer16,43,41 have been demonstrated to predispose participants to higher

rates of injury compared to physical exercise of a non-contact nature61, such as volleyball45, in

which the forces exerted on the participants’ joints, and thus the incidence, prevalence and

severity of injury, may differ considerably62. Therefore, whilst a variety of physical exercise

types were required to meet the objectives of this systematic review, the comparability of the

results may have decreased as a result.

Reporting of Results

The reporting of results was generally of high quality across studies, with clear and in-depth

discussions of their findings, and probability values set at the commonly accepted level of

p≤0.05 or less, facilitating comparison63. However, three studies16,39,42 failed to explicitly

explore their study limitations, frustrating any assessment of author bias.

Strengths and Limitations of this Review

16

Nine studies were included in this review, which is only half the number of papers included

in the review by Pacey et al.21. However, six of the studies were published after Pacey et al.21

and our review therefore drew primarily from more recent research to formulate its findings.

The current review also aimed to investigate three categories of outcome (incidence,

prevalence and severity) and both upper and lower limb injuries to further explore the links

between GJL and limb injury related to physical exercise. However, the studies retrieved

predominantly explored injury incidence and prevalence, and lower limb injuries. Thus, a

sufficiently conclusive review of injury severity or upper limb injuries could not be

established. This review specifically excluded those with arthralgia associated with

recognised connective tissue disorders (such as JHS or Ehlers-Danlos Syndrome). The

findings are therefore relevant to a more general population with GJL.

Recommendations for Future Research

Based on the limitations of this review, it is recommended that future research should further

investigate the links between GJL and upper limb injury and injury severity. Furthermore,

future research may benefit from utilising prospective cohort study designs that adequately

account for potential confounders in their design and analysis. It is recommended that a more

diverse range of physical activities and participant ages should be included, and that study

designs should incorporate blinding and prospective sample size calculations.

Conclusion

Overall, this review supports links between GJL and lower limb injury incidence and

prevalence, across numerous forms of physical exercise. However, evidence for potential

links between GJL and upper limb injury, or injury severity, was inconclusive. Therefore,

future primary research is advocated to further investigate the link between GJL and upper

17

limb injury, and limb injury severity, via high-quality prospective cohort studies. This would

provide a more holistic impression of the impact of GJL among populations whom undertake

regular physical exercise.

Declaration of Interest Statement: All authors can confirm that no financial interest or

benefit has arisen from the research.

18

References

1. Quatman, C.E., Forda, K.R., Myera, G.D., Paternoa, M.V. and Hewett, T.E. The effects of

gender and pubertal status on generalized joint laxity in young athletes. J Sci Med Sport.

2008 Jun; 11(3):257-63.

2. Kim, S.J., Kumar, P. and Kim, S.H. Anterior cruciate ligament reconstruction in patients

with generalized joint laxity. Clin Orthop Surg. 2010 Aug; 2(3):130-139.

3. Pearsall, A.W., Kovaleski, J.E., Heitman, R.J., Gurchiek, L.R. and Hollis, J.M. The

relationships between instrumented measurements of ankle and knee ligamentous laxity and

generalized joint laxity. J Sports Med Phys Fitness. 2006 Mar; 46(1):104-110.

4. Boyle, K.L., Witt, P. and Riegger-Krugh, C. Intrarater and interrater reliability of the

Beighton and Horan joint mobility index. J Athl Train. 2003 Oct-Dec; 38(4):281–285.

5. Birrell, F.N., Adebajo, A.O., Hazleman, B.L. and Silman, A.J. High prevalence of joint

laxity in West Africans. Br J Rheumatol. 1994 Jan; 33(1):56-59

6. Beighton, P., Solomon, L. and Soskolne, C.L. Articular mobility in an African population.

Ann Rheum Dis, 1973 Sep; 32(5):413–418.

7. Vallis, A., Wray, A. and Smith, T. Inter- and Intra-rater Reliabilities of the Beighton Score

Compared to the Contompasis Score to Assess Generalised Joint Hypermobility. MYOPAIN.

2015 Apr; 23(1-2):21-27

8. Hirsch, C., Hirsch, M., John, M.T. and Bock, J.J. Reliability of the Beighton

Hypermobility Index to Determinate the General Joint Laxity Performed by Dentists. J

Orofac Orthop 2007, 68(5):342–352.

9. Grahame, R. Hypermobility – not a circus act. Int J Clin Pract. Jun 2000, 54(5):314-315

19

10. Hauser, R.A. and Phillips, H.J. Treatment of Joint Hypermobility Syndrome, Including

Ehlers-danlos Syndrome, with Hackett-hemwall Prolotherapy. J Prolotherapy 2011 May;

3(2):612-629

11. Pacey, V., Tofts, L., Wesley, A., Collins, F. and Singh-grewal, D. Joint Hypermobility

Syndrome: A Review For Clinicians. J Paediatr Child Health. 2015 Apr; 51(4):373-380

12. Hakim, A. and Grahame, R. Joint Hypermobility. Best Pract Res Clin Rheumatol. 2003

Dec; 17(6):989-1004

13. Juul-Kristensen, B., Røgind, H., Jensen, D.V. and Remvig, L. Inter-examiner

Reproducibility of Tests and Criteria For Generalized Joint Hypermobility and Benign Joint

Hypermobility Syndrome. Rheumatology. 2007 Dec; 46(12):1835-1841

14. Malfait, F., Francomano, C., Byers, P., Belmont, J., Berglund, B., Black, J., Bloom, L.,

Bowen, J.M., Brady, A.F., Burrows, N.P., Castori, M., Cohen, H., Colombi, M., Demirdas,

S., De Backer, J., De Paepe, A., Fournel-Gigleux, S., Frank, M., Ghali, N., Giunta, C.,

Grahame, R., Hakim, A., Jeunemaitre, X., Johnson, D., Juul-Kristensen, B., Kapferer-

Seebacher, I., Kazkaz, H., Kosho, T., Lavallee, M.E., Levy, H., Mendoza-Londono, R.,

Pepin, M., Pope, F.M., Reinstein, E., Robert, L., Rohrbach, M., Sanders, L., Sobey, .GJ., Van

Damme, T., Vandersteen, A., van Mourik, C., Voermans, N., Wheeldon, N., Zschocke, J.,

Tinkle, B. The 2017 international classification of the Ehlers–Danlos syndromes. Am J Med

Genet Part C Semin Med Genet. 2017; 175C:8-26.

15. Scheper, M., De Vries, J., Verbunt, J., Nollet, F. and Engelbert, R. FRI0455 Is the

presence of generalized joint hypermobility in young adult female dancers beneficial?. Ann

Rheum Dis. 2013 Jun; 71(3),468

20

16. Stewart, D.R, Burden, S.B. Does generalised ligamentous laxity increase seasonal

incidence of injuries in male first division club rugby players? Br J Sports Med. 2004 Aug;

38(4):457-460

17. ÖStenberg, A. and Roos, H. Injury risk factors in female European football. a prospective

study of 123 players during one season. Scand J Med Sci Sports. 2000 Oct; 10(5):279-285

18. Myer, G.D., Ford, K.R., Paterno, M.V., Nick, T.G. and Hewett, T.E. The effects of

generalized joint laxity on risk of anterior cruciate ligament injury in young female athletes.

Am J Sports Med. 2008 Jun; 36(6):1073–1080

19. Beynnon, B.D., Renström, P.A., Alosa, D.M., Baumhauer, J.F. and Vacek, P.M. Ankle

ligament injury risk factors: a prospective study of college athletes. J Orthop Ress 2001 Mar;

19(2):213-20

20. McCormack, M., Briggs, J., Hakim, A. and Grahame, R. Joint laxity and the benign joint

hypermobility syndrome in student and professional ballet dancers. J Rheumatol. 2004 Jan;

31(1):173-178

21. Pacey, V., Nicholson, L.L., Adams, R.D., Munn, J. and Munns, C.F. Generalised Joint

Hypermobility and Risk of Lower Limb Joint Injury During Sport: A Systematic Review

with Meta-Analysis. Am J Sports Med. 2010 Jul; 38(7):1487-1497

22. Söderman, K., Alfredson, H., Pietilä, T. and Werner, S. Risk factors for leg injuries in

female soccer players: a prospective investigation during one out-door season. Knee Surg

Sports Traumatol Arthrosc. 2001 Sep; 9(5):313-321

23. Hiller, C.E., Refshauge, K.M., Herbert, R.D. and Kilbreath, S.L. Intrinsic predictors of

lateral ankle sprain in adolescent dancers: a prospective cohort study. Clin J Sport Med. 2008

Jan; 18(1):44-48

21

24. Uhorchak JM, Scoville CR, Williams GN, Arcerio RA, St Pierre P, Taylor DC. Risk

factors associated with noncontact injury of the anterior cruciate ligament: a prospective four-

year evaluation of 859 West Point cadets. Am J Sports Med. 2003 Nov-Dec; 31(6):831-842

25. Watson, R.R. Quantitative research. Nurs Stand. 2015 Apr; 29(31):44-48

26. Saremi, H., Yavarikia, A. and Jafari, N. Generalized Ligamentous Laxity: An Important

Predisposing Factor for Shoulder Injuries in Athletes. Iran Red Crescent Med J. 2016 Jun;

18(6):1-5

27. van Mechelen W., Hlobil H., Kemper H.C. Incidence, Severity, Aetiology and Prevention

of Sports Injuries. a Review of Concepts. Sports Med. 1992 Aug; 14(2):82-99

28. Doherty, C., Delahunt, E., Caulfield, B., Hertel, J., Ryan, J. and Bleakley, C. The

Incidence and Prevalence of Ankle Sprain Injury: A Systematic Review and Meta-analysis of

Prospective Epidemiological Studies. Sports Med. 2014 Jan; 44(1):123-140

29. World Health Organisation (WHO). Physical Activity 2016.

http://www.who.int/topics/physical_activity/en/ Accessed December 10, 2016

30. Caspersen, C.J., Powell, K.E. and Christenson, GM. Physical Activity, Exercise, and

Physical Fitness: Definitions and Distinctions For Health-related Research. Public Health

Rep. 1985 Mar-Apr; 100(2):126–131

31. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).

Transparent reporting of systematic reviews and meta-analysis 2016. http://www.prisma-

statement.org Accessed December 08, 2016

32. Aromataris, E. and Pearson, A. The systematic review: an overview. Am J Nurs 2014

Mar; 114(3):53-58

33. Davis, D. A practical overview of how to conduct a systematic review. Nurs Stand. 2016

Nov 16; 31(12):60-71

22

34. Lai, N.M. Dissecting students' bedside clinical questions using the 'pico' framework. Med

Educ 2009 Apr; 43(5):479–480

35. Jahan, N., Naveed, S., Zeshan, M. and Tahir, M.A. How to conduct a systematic review: a

narrative literature review. Cureus. 2016 Nov; 8(11):1-8

36. Clarke, S. Collier. S. Research essentials. How to critique quantitative research. Nurs

Child Young People. 2015 Nov; 27(9):12

37. Hek, G. & Moule, P. Making Sense of Research. 3rd Ed. London: Sage Publications; 2006

38. Crowe, M. and Sheppard, L. A general critical appraisal tool: an evaluation of construct

validity. Int J Nurs Stud. 2011 Dec; 48:1505-1516.

39. Akodu, A.K., Aiyegbusi, A.I. and Agbaje, T.R. Prevalence of generalized joint

hypermobility and its association with sports injuries at recreational cricket players. Medicina

Sportiva: J Romanian Sport Medicine Society 2016 Jan; 12(1),2727-2732

40. Bin Abd Razak, H.R., Ali, N.B. and Howe, T.S. Generalized ligamentous laxity may be a

predisposing factor for musculoskeletal injuries. J Sci Med Sport. 2014 Sep; 17(5):474-8

41. Blokland, D., Thijs, K.M., Backx, F.J., Goedhart, E.A. and Huisstede, B.M. No Effect of

Generalized Joint Hypermobility on Injury Risk in Elite Female Soccer Players: A

Prospective Cohort Study. Am J Sports Med. 2017 Feb; 45(2):286-293

42. Decoster, L.C., Bernier, J.N., Lindsay, R.H. and Vailas, J.C. Generalized joint

hypermobility and its relationship to injury patterns among NCAA lacrosse players. J Athl

Train. 1999 Apr-Jun; 34(2):99–105.

43. Konopinski, M.D., Jones, G.J. and Johnson, M.I. The Effect of Hypermobility on the

Incidence of Injuries in Elite-Level Professional Soccer Players A Cohort Study. Am J Sports

Med. 2012 Dec; 40:763–769.

23

44. Smith, R., Damodaran, A.K., Swaminathan, S., Campbell, R. and Barnsley, L.

Hypermobility and sports injuries in junior netball players. Br J Sports Med. 2005 Sep;

39(9):628-31

45. Sueyoshi, T., Emoto, G. and Yuasa, T. Generalized Joint Laxity and Ligament Injuries in

High School–Aged Female Volleyball Players in Japan. Orthop J Sports Med. 2016 Oct;

4(10):1-5

46. Song, J.W. and Chung, K.C. Observational studies: cohort and case-control studies. Plast

Reconstr Surg. 2010 Dec; 126(6):2234–2242

47. Hollands, H., Brox, A.C. and Kertes, P.J. Sample Size Calculation. E B Ophthalmology.

2008 Jul; 9(3):153-157.

48. Nayak, B.K. Understanding the relevance of sample size calculation. Indian J

Ophthalmol. 2010 Nov-Dec; 58(6):469-470

49. Gogtay, N. Principles of sample size calculation. Indian J Ophthalmol. 2010 Nov-Dec;

58(6):517-518

50. Boffetta, P. Internal and external validity of cohort studies. Ann Agric Environ Med

2011;18(2):283-284

51. Elwood, J.M. Critical Appraisal of Epidemiological Studies and Clinical Trials. 3rd ed.

Oxford: Oxford University Press 2007

52. Altman, D.G., Schulz, K.F., Moher, D., Egger, M., Davidoff, F., Elbourne, D., Gã¸tzsche,

P.C. et al. Revised consort statement for reporting randomized trials: explanation and

elaboration. Ann Intern Med. 2001 Apr; 134(8):663-694

53. Bookwala, A., Hussain, N. and Bhandari, M. The three-minute appraisal of a prospective

cohort study. Indian J Orthop 2011 Jul-Aug; 45(4): 291–293.

24

54. Delgado-Rodríguez, M. and Llorca, J. Bias. J Epidemiol Community Health 2004 Aug;

58(8):635–641

55. Gefeller, O. Invited commentary: recall bias in melanoma-much ado about almost

nothing?. Am J Epidemiol 2009 Nov; 169(3):267–270

56. Porritt, K., Gomersall, J. and Lockwood, C. Study Selection and Critical Appraisal. Am J

Nurs. 2014 Jun; 114(6):47-52.

57. Little, J., Bradley, L., Bray, M.S., Clyne, M., Dorman, J., Ellsworth, D.L., Hanson, J., et

al. Reporting, appraising, and integrating data on genotype prevalence and gene-disease

associations. Am J Epidemiol. 2002 Aug; 156(4),300-310

58. Fuller, C.W., Ekstrand, J., Junge, A., Andersen, T.E., Bahr, R., Dvorak, J., Hägglund, M.,

et al. Consensus Statement on Injury Definitions and Data Collection Procedures in Studies

of Football (Soccer) Injuries. Br J Sports Med. 2006 Mar; 40(3):193–201.

59. Woodward, M. Epidemiology: Study Design and Data Analysis. 3rd ed. Boca Raton:

Taylor & Francis 2014

60. Holland, G.J., Tanaka, K., Shigematsu, R. and Nakagaichi, M. Flexibility and physical

functions older adults: a review. J Aging Phys Act 2002 Apr; 10:169-206.

61. Hootman, J.M., Dick, R. and Agel, J. Epidemiology of collegiate injuries for 15 sports:

summary and recommendations for injury prevention initiatives. J Athl Train. 2007 Apr-Jun;

42(2):311–319.

62. Bahr, R. and Krosshaug, T. Understanding injury mechanisms: a key component of

preventing injuries in sport. Br J Sports Med. 2005 Jun; 39(6):324-9.

63. Barton, B. and Peat, J.K. Medical Statistics: A Guide to SPSS, Data Analysis, and Critical

Appraisal. Chichester: Wiley Blackwell 2014.

25

TABLES

Table 1. Search Terms. * = word truncation.

Search 1 (Population) Search 2 (Exposure) Search 3 (Outcome)

Generali* Joint Laxity Physical Exercis* Limb

Generali* Joint Hypermobility Sport* Injur*

Generali* Joint Hyper-mobile Physical Activit* Beighton

Ligament* Laxity Exercis*

Table 2. Inclusion/ Exclusion criteria. GJL = Generalized Joint Laxity.

Inclusion Criteria Exclusion Criteria

English Language. Language other than English.

Human participants. Non-human participants.

Identification of GJL in participants. Connective tissue disorders other than GJL.

Beighton score used to identify GJL. A scoring system other than the Beighton

score used to identify GJL.

Participants involved in physical exercise. Participants not involved in physical

exercise.

Reports data related to limb injury. No report of limb injury data.

Reports incidence, prevalence and/or

severity of limb injury.

No relevant outcomes reported.

Reports primary research data. Secondary research.

Predominantly designed to investigate the

links between GJL and limb injury.

The links between GJL and limb injury

investigated as a secondary aim.

26

Table 3. Study characteristics. GJH = Generalized Joint Hypermobility; GJL = Generalized Joint Laxity; NCAA = National Collegiate Athletic Association.Characteristics Akodu et al. (2016)39 Bin Abd Razak et

al. (2014)40Blokland et al.

(2017)41Decoster et al.

(1999)42Konopinski et al.

(2012)43

Study Design Cohort Study (Prospective)

Case-control Study (Prospective)

Cohort Study (Prospective)

Cohort Study (Prospective)

Cohort Study (Prospective)

Country Nigeria Singapore Netherlands United States of America

United Kingdom

Aims ‘…investigating the prevalence of [GJH] and its association with sports injuries in recreational cricket players.’

‘…determine if [GJL] may be a predisposing factor for musculoskeletal injuries in young males.’

‘…prospectively investigate whether GJH is a risk factor for soccer injuries in elite female soccer players.’

‘…prospectively observe injury patterns among hypermobile and nonhypermobile athletes over one athletic season.’

‘…compare the incidence, severity, location, and nature of injuries in hypermobile and nonhypermobile professional soccer players…’

Limbs investigated Upper and lower limb

Upper and lower limb

Upper and lower limb (only lower limb data is clearly identifiable)

Upper and lower limb (only lower limb data is clearly identifiable)

Upper and lower limb data (only lower limb data is clearly identifiable)

Beighton cut-off score (max 9)

Moderately hypermobile 3-4; Distinctly hypermobile 5-9

≥4 ≥4 ≥5 ≥4

Sample Size (n) 102[16 not hypermobile; 35 moderately hypermobile; 51 distinctly hypermobile]

200[100 cases with musculoskeletal injuries; 100 controls with no musculoskeletal

114[20 hypermobile]

310[147 males (20 hypermobile): 163 females (54 hypermobile)]

54[18 hypermobile]

27

injuries]Sex All male All male All female Mixed All maleAge (years) Range 18-38

Mean 23.3Range 18-25Mean 22.3 (cases) Mean 22.4 (controls)

Range 17.1-33.0Mean 22.4

Mean 20 Mean 22.5

Physical Exercise Cricket Military training Soccer Lacrosse SoccerOutcome Categories

Injury prevalence Injury prevalence Injury incidence, injury severity

Injury incidence, injury severity

Injury incidence, injury prevalence, injury severity

Main Statistical Findings

No significant association between GJL and upper limb injury prevalence (r=-0.187, p=0.061). Significant association between GJL and lower limb injury prevalence (r=-0.250, p=0.011).

Cases were more likely to have GJL (n=12, 12%) compared to controls (n=4, 4%) (odds ratio 3.35, p=0.043).

No significant differences between hypermobile and non-hypermobile players in the number of injuries per player (p=0.382); injury incidence (p=0.551); injury location (all body areas p>0.05); or injury severity (days to full recover) (all p>0.05). No significant difference in incidence rate ratios (p=0.662) or odds ratios (p=0.520) for all injuries.

No significant difference in injury incidence between hypermobile and non-hypermobile athletes (p=0.18). No difference in injury severity (time lost). No difference in occurrence of sprains, fractures, bursitis or cartilage injuries. Hypermobile athletes had an increased incidence of ankle injury (p<0.05) and reduced incidence of contact injuries (p=0.037).

Hypermobile players had a greater total number of injuries (p<0.001); higher injury incidence (p<0.001); higher total days missed because of injury (p<0.001); and higher incidence of reinjury (p<0.001). Hypermobile players had more mild (p=0.011), moderate (p=0.02) and severe injuries (p<0.001). The location and type of injury did not differ.

28

Table 3 (continued). Study characteristics. GJL = Generalized Joint Laxity.

Characteristics Saremi et al. (2016)26 Smith et al. (2005)44

Stewart & Burden (2004)16

Sueyoshi et al. (2016)45

Study Design Cross-sectional Study (Retrospective)

Cross-sectional Study(Retrospective)

Cohort Study(Prospective)

Case control study(Retrospective)

Country Iran Australia New Zealand JapanAims ‘…determine whether

[GJL] can be a predisposing factor for acute and chronic shoulder injuries in athletes.’

‘…investigate the association between joint mobility and injuries in netball players.’

‘…investigate if ligamentous laxity increases seasonal incidence of injury in male first division club rugby players...’

‘…investigate [GJL] and a history of ligament injury in high school-aged volleyball players.’

Limbs investigated Upper limb (shoulder) Upper and lower limb Upper and lower limb Upper and lower limb (only lower limb injuries were reported)

Beighton cut-off score (max 9)

≥4 Moderately hypermobile 3-4; Distinctly hypermobile 5-9

Hypermobile 4-6; Extremely hypermobile 7-9

No cut-off. Mean score used to compare groups

Sample Size (n) 118 [80 males: 38 females; 43 hypermobile]

200[70 not hypermobile; 51 moderately hypermobile; 79 distinctly hypermobile]

51[39 not hypermobile; 8 hypermobile; 4 extremely hypermobile]

47[30 with ligament injury history (19 single injury, 11 multiple injury); 17 with no ligament injury history]

Sex Mixed All female All male All femaleAge (years) Range 17-37 Range 6-16

Mean 11Mean 23.6 Range 13-18

Mean 15.83 (injury)Mean 16.13 (non-injury)

Physical Exercise Basketball, handball, volleyball, football,

Netball First-division rugby Volleyball

29

swimming, water-polo, mountaineering, climbing, fitness, martial arts, boxing, racket sports, boating, wrestling

Outcome Categories Injury prevalence, injury severity

Injury prevalence Injury incidence Injury prevalence

Main Statistical Findings

The GJL group had greater prevalence of chronic shoulder pain (p=0.016); more chronic shoulder injuries (p=0.032); more shoulder instability (p=0.004); and less functionality (p=0.030). There was no difference in the prevalence of acute shoulder injuries (p=0.58).

21% of non-hypermobile children had sustained a netball injury compared with 37% of moderately hypermobile and 43% of distinctly hypermobile. These differences were statistically significant (p<0.025). Beighton score was an independent risk factor for netball injury (p=0.017). The odds ratios for netball injury were 3.364 for moderately hypermobile (p=0.015) and 2.998 for distinctly hypermobile (p=0.010) children when compared to non-hypermobile children.

No significant differences in injury rate between the three laxity groups (p<0.05). When the two hypermobility groups were combined, injury incidence was significantly higher in hypermobile (116.7/1000 hours) than tight (43.6/1000 hours) players (p=0.035).

The injury group had higher mean Beighton scores (2.40/9) than the non-injury (1.24/9) group (p=0.006). The multiple injury group had higher mean Beighton scores (3.18/9) than the single injury (1.95/9) group (p=0.02).

30

Table 4a. SIGN cohort checklist. = Yes; = No; N/A = Not Applicable; ? = Can’t Say; ++ = High quality; + = Acceptable.

Criteria Akodu et al

(2016)39

Blokland et al.

(2017)41

Decoster et al.

(1999)42

Konopinski et al.

(2012)43

Saremi et al.

(2016)26

Smith et al.

(2005)44

Stewart & Burden (2004)16

1.1. The study addresses an appropriate and clearly focused question.

1.2. The two groups being studied are selected from source populations that are comparable in all respects other than the factor under investigation.

1.3. The study indicates how many of the people asked to take part did so, in each of the groups being studied.

1.4. The likelihood that some eligible subjects might have the outcome at the time of enrolment is assessed and taken into account in the analysis.

N/A N/A

1.5. What percentage of individuals or clusters recruited into each arm of the study dropped out before the study was completed.

0% 8.8% 0% 3.7% 0% 0% 0%

1.6. Comparison is made between full participants and those lost to follow up, by exposure status.

N/A N/A N/A N/A N/A

1.7. The outcomes are clearly defined.

1.8. The assessment of outcome is made blind to exposure status. If the study is retrospective this may not be applicable.

1.9. Where blinding was not possible, there is some recognition that

31

knowledge of exposure status could have influenced the assessment of outcome.1.10. The method of assessment of exposure is reliable.

1.11. Evidence from other sources is used to demonstrate that the method of outcome assessment is valid and reliable.

?

1.12. Exposure level or prognostic factor is assessed more than once.

N/A

1.13. The main potential confounders are identified and taken into account in the design and analysis.

? ?

1.14. Have confidence intervals been provided?

2.1. How well was the study done to minimise the risk of bias or confounding?

+ ++ + ++ + + +

2.2. Taking into account clinical considerations, your evaluation of the methodology used, and the statistical power of the study, do you think there is clear evidence of an association between exposure and outcome?

2.3. Are the results of this study directly applicable to the patient group targeted in this guideline?

32

33

Table 4b. SIGN case-control checklist. = Yes; = No; N/A = Not Applicable; ? = Can’t Say. ++ = High quality; + = Acceptable.

34

Criteria Bin Abd Razak et al. (2014)40

Sueyoshi et al. (2016)45

1.1 The study addresses an appropriate and clearly focused question 1.2 The cases and controls are taken from comparable populations 1.3 The same exclusion criteria are used for both cases and controls ?1.4 What percentage of each group (cases and controls) participated in the study? Case 50% Case 64%

Control 50% Control 36%1.5 Comparison is made between participants and non-participants to establish their similarities or differences

1.6 Cases are clearly defined and differentiated from controls 1.7 It is clearly established that controls are non-cases 1.8 Measures will have been taken to prevent knowledge of primary exposure influencing case ascertainment

1.9 Exposure status is measured in a standard, valid and reliable way 1.10 The main potential confounders are identified and taken into account in the design and analysis

1.11 Confidence intervals are provided 2.1 How well was the study done to minimise the risk of bias or confounding? ++ +2.2 Taking into account clinical considerations, your evaluation of the methodology used, and the statistical power of the study, do you think there is clear evidence of an association between exposure and outcome?

2.3 Are the results of this study directly applicable to the patient group targeted by this guideline?

Figure 1. Online search inclusion/exclusion flow diagram. Adapted from PRISMA31.

35

Potential studies identified(n = 421)

Potential studies identified through snowballing

(n = 25)

Potential studies identified through database searching

(n = 396)

Studies excluded based on secondary aim (n = 9)

Studies screened by primary or secondary aim (n = 18)

Studies excluded based on full text (n = 4)

Duplicates removed(n = 147)

Studies included for critical appraisal (n = 9)

Studies screened by full text (n = 22)

Studies excluded based on abstract (n = 92)

Studies screened by abstract(n = 114)

Studies excluded based on title (n = 160)

Studies screened by title(n = 274)

Eligibility

Identification

Included

Screening

TABLE AND FIGURE HEADINGS

Table 1. Search Terms. * = word truncation.

Table 2. Inclusion/ Exclusion criteria. GJL = Generalized Joint Laxity.

Table 3. Study characteristics. GJH = Generalized Joint Hypermobility; GJL = Generalized

Joint Laxity; NCAA = National Collegiate Athletic Association.

Table 3 (continued). Study characteristics. GJL = Generalized Joint Laxity.

Table 4a. SIGN cohort checklist. = Yes; = No; N/A = Not Applicable; ? = Can’t Say;

++ = High quality; + = Acceptable.

Table 4b. SIGN case-control checklist. = Yes; = No; N/A = Not Applicable; ? = Can’t

Say. ++ = High quality; + = Acceptable.

Figure 1. Online search inclusion/exclusion flow diagram. Adapted from PRISMA31.

36