Physical Therapy in Sport 10 (2009) 91–96
Contents lists available at ScienceDirect
Physical Therapy in Sport
journal homepage: www.elsevier.com/ptsp
Original research
Hypermobility, injury rate and rehabilitation in a professional football
squad – A preliminary study
Richard Collinge a, *, Jane V. Simmonds b
a
b
UCL Training Ground, London Colney, Hertfordshire AL2 1BZ, UK
School of Health & Emergency Professions, University of Hertfordshire, Hatfield AL10 9AB, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 4 December 2008
Received in revised form
6 February 2009
Accepted 3 March 2009
Objectives: To determine if joint hypermobility is a risk factor for injury in a professional football squad.
Primary objectives were to estimate the prevalence of hypermobility amongst a professional football
squad and to undertake an audit of injuries sustained over a season. Secondary objectives were to relate
the injury audit findings and hypermobility levels to time missed through injury, assessed by training
days and competitive first team games missed after musculo-skeletal injury.
Hypothesis: Increasing levels of joint hypermobility may result in an increased risk of injury in a contact
sport such as professional football.
Design: A prospective observational study consisting of the Beighton joint hypermobility screen and an
injury audit (season 2007/8).
Setting: A second tier, English professional football club.
Keywords:
Joint hypermobility
Laxity
Beighton scale
Professional football
Injury audit
Return to play timeframes
Participants: Thirty-three male professional footballers aged 18–35 years.
Main outcome measures: The Beighton joint hypermobility screen and a seasonal injury audit.
Results: The prevalence of joint hypermobility was found to be between 21 and 42% depending on the
cut-off score used for the Beighton scale. Similar injury rates were found in both the hypermobile and
non-hypermobile participants (6.2 as compared to 6.3 injuries/1000 h exposure respectively). Once
injured, the hypermobile group showed a tendency towards missing more competitive first team games
(12 as compared to 5/season in non-hypermobiles) and training days (71 as compared to 31 days/season
in non-hypermobiles). These findings were not statistically significant.
Conclusions: The prevalence of joint hypermobility in a cohort of professional footballers is comparable
to previous studies in athletic populations and is dependent upon which Beighton cut-off score is
selected. It may be inferred from this preliminary study that the return to play timescales in hypermobile
individuals may be extended so as to minimise the potential risk of re-injury and limit the socioeconomic
costs associated with time out of competition.
Ó 2009 Elsevier Ltd. All rights reserved.
1. Introduction
Hypermobile joints are described as those displaying a range of
movement that is considered excessive, taking into consideration
the age, gender and ethnic background of the individual (Grahame,
2003). Joint Hypermobility Syndrome (JHS) is present when an
individual presents with hypermobile joints in the absence of
demonstrable rheumatological disease (Grahame, 2003). Where
once hypermobility was thought to be uncommon, it is now
classified as a hereditary connective tissue disorder (HCTD), sharing
common features with more serious counterparts (Fig. 1).
* Corresponding author. Tel.: þ44 07974772014; fax: þ44 01727821211.
E-mail address: docsamanthony@yahoo.co.uk (R. Collinge).
1466-853X/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ptsp.2009.03.001
The epidemiology of joint hypermobility (JH) and JHS varies in
the literature reviewed (Simmonds & Keer, 2007), with females and
individuals from Asian and African backgrounds being more prone
to the presentation (Russek, 1999). The prevalence of JH and the
syndrome in the adult population has been reported at between 10
and 30% (Hakim & Grahame, 2003). Joint laxity is usually greatest at
birth, decreases during childhood and continues to reduce during
adolescence and adult life (Middleditch, 2003), suggesting that
there are age related changes in flexibility levels.
The underlying pathophysiology in JH is attributed to imbalances in the proportions of different forms of collagen, extracellular proteins or hormonal factors that affect the soft tissue
matrix. A strong familial aggregation has been linked to the HCTDs
and an autosomal dominant mode (Malfait, Hakim, De Paepe, &
92
R. Collinge, J.V. Simmonds / Physical Therapy in Sport 10 (2009) 91–96
Ehlers-Danlos
Marfan
Classic vascular types
Aortic dilatation
Ectopia lentis
Marfanoid
habitus
Stretchy skin,
Papyraceous
scars
Joint Hypermobility
Skin striae
(Syndrome)
Osteopenia
Osteoporosis
Fractures
Osteogenesis
Imperfecta
Fig. 1. A flow diagram showing the relationship between joint hypermobility
(syndrome) and the heritable disorders of connective tissue (adapted from Grahame,
2003).
Grahame, 2006). The manifestation of this genetic encoding is the
subsequent effect upon the connective tissue matrix; namely the
collagens, fibrillins, elastins and proteoglycans. It is the make up of
the connective tissue matrix that determines an individual’s flexibility, as well as joint capsule, ligament and tendon mechanical
properties.
It has been suggested that it is the abnormal ratio of type III:type
I collagen that results in the decreased tissue stiffness common to
JH patients, with the thin and elastic type III collagen becoming
more prevalent within the soft tissue matrix (Russek, 1999). Tissue
biopsies have supported the hypothesis that it is the interference
with the processing of the N-propeptide of either a-chain (a1 or a2)
of type I collagen or mutations in the production of type V collagen
that result in hypermobile presentations (Malfait et al., 2006).
Another mutation in a non-collagenous molecule, called tenascinX, has also been suggested as a predisposing factor towards JH
(Schalkwijk et al., 2001; Zweers, Dean, van Kuppevelt, Bristow, &
Schalkwijk, 2005), whilst elevated levels of IGF-1, insulin and
growth hormone have also been discovered in JHS patients (Denko
& Boja, 2001).
It has been suggested that proprioceptive acuity is reduced in
hypermobile subjects. This has been reported at the knee (Hall,
Ferrell, Sturrock, Hamblen, & Baxendale, 1995) and proximal interphalangeal joints of the finger (Mallik, Ferrell, McDonald, & Sturrock, 1994). It has also been suggested that joint hypermobility may
affect other afferent apparatus of the nervous system and subsequently lead to chronic, sensitised pain states (Grahame, 2003) and
in the longer term, osteoarthritis (Al Rawi & Nessan, 1997; Jonsson,
Valtysdottir, Kjartansson, & Brekkan, 1996).
At present, the optimum level of flexibility required to prevent
injury when participating in sporting activity is not clear and may
vary between muscle groups and probably sports (Dadebo, White,
& George, 2004). A knowledge of where athletes fall in the spectrum of joint mobility may influence intervention and understanding of their complaints (Boyle, Witt, & Riegger-Krugh, 2003)
and it has been postulated that athletes at either end of the flexibility spectrum are likely to be more at risk of injury (Stewart &
Burden, 2004). Indeed, individuals presenting with generalised
joint laxity have been reported to have an increased risk of Anterior
Cruciate Ligament (ACL) injury to the knee (Ramesh, Von Arx,
Azzopardi, & Schranz, 2005), with the risk increasing five fold in
female football players who demonstrate hyperextension postures
of the knee (Myer, Ford, Paterno, Nick, & Hewett, 2008).
The implications for hypermobile subjects competing in sport
have produced conflicting results in the literature, mainly due to
the diversity of both contact and non-contact sports studied but
also due to inconsistencies in study design. Hypermobile athletes
have been shown to have an increased risk of injury in American
football (Nicholas, 1970), amateur rugby (Stewart & Burden, 2004)
and netball (Smith, Damodaran, Swaminathan, Campbell, &
Barnsley, 2005), whilst contradictory evidence has been published
for lacrosse participants (Decoster, Bernier, Lindsay, & Vailas, 1999)
and American Collegiate athletes (Krivickas & Feinberg, 1996). To
date, no studies have examined whether hypermobility is a risk
factor for injury in a professional football squad.
On a more general level, Hardin, Voight, Blackburn, Canner, and
Soffer (1997) report a slower rehabilitation course for individuals
presenting with joint hypermobility, highlighting the need for
effective screening and preventative programmes to manage
hypermobile athletes and limit the socioeconomic costs associated
with sports injuries.
The primary objective and hypothesis of this research was to
estimate the prevalence of JH in a professional football squad and
to determine if increasing levels of joint hypermobility, measured
using the Beighton scale (Beighton, Soskolne, & Solomon, 1973),
results in an increased risk of injury in a professional football
squad.
The issue of the cut-off score that denotes hypermobility diagnosis has caused much confusion in the hypermobility literature.
There is no universal agreement on a threshold for JH; some
researchers use a Beighton score of 4 or 5/9, other researchers use
a Beighton scale score of 6/9 and still other researchers use
a modified score of 3/5 (Russek, 1999). The latter is scored out of 5
by combining the scores obtained by the left and right peripheral
joint tests. For the purposes of this study, the cut-off scores utilised
by Stewart and Burden (2004) in a study of rugby players, where
a score of 4 or greater represents hypermobility and 7 or greater
represents excessive hypermobility, will be considered. The cut-off
score of 4 points also correlates with the criteria used in the Revised
1998 Brighton Criteria (Grahame, Bird, & Child, 2000). Boyle et al.
(2003) found the Beighton scale to have good to excellent reliability
in screening individuals aged 15–45 years, with intra-rater reliability reported at 81% when a cut-off score of 5 or greater was
selected. This scoring classification will also be referred to in this
study.
Secondary objectives of this study were to relate the seasonal
injury audit findings and hypermobility levels to time missed
through injury, assessed by training days and competitive first team
games missed after musculo-skeletal injury.
2. Methods
2.1. Participants
Thirty-three professional footballers signed to a second tier
English team for the 2007/8 season volunteered for the study. The
participants were aged 18–35 years and had all signed professional
contracts. All players were available to train on a full time basis and
were eligible for selection for first team fixtures. Demographic
information for the participant group can be seen in Table 1.
When analysed by player position, the breakdown of the
participant group consisted of three goalkeepers, eleven defenders,
nine midfielders and ten attackers (Fig. 2). Three of the players
were carrying injuries over from the 2006/7 season (two players
had undergone ACL reconstructions and the other internal fixation
of a tibial fracture) and were classified as having been injured at the
start of the study on 1st July 2007.
R. Collinge, J.V. Simmonds / Physical Therapy in Sport 10 (2009) 91–96
93
Table 1
A table representing demographic data of the participant group (one standard deviation, data range).
Participant number
Mean age – years
Mean career length – years
Mean height – cm
Mean weight – kg
Lower limb dominance
33
24.4 (4.8, 18–35 years)
7 (4.8, 1–17 years)
179.2 (6.2, 165–192.5 cm)
82.6 (9, 67.3–109.6 kg)
Left ¼ 6
Right ¼ 27
Ethical approval was obtained from the University College
London ethics committee. Informed consent was obtained from all
participants prior to entering the study.
2.2. Study design
A prospective observational study design was undertaken.
Participants were invited to take part in a 20 min assessment
session with the principal investigator (RC) in August 2007, incorporating the Beighton scale as a measure of hypermobility status
(Table 2).
An anonymous questionnaire was also completed, with participants allocated individual identification numbers, which included
ethnic background, age, playing position, career length as
a professional footballer, lower limb dominance, height and weight.
Lower limb dominance was defined as the leg that the participant
would strike a ball with instinctively for maximal distance. Details
of previous and/or current injury that may have affected any
component of the test procedure were documented.
Based upon the classification used by Stewart and Burden
(2004), which was based on rugby players, the participant’s
Beighton scores were analysed in accordance with the criteria
shown in Table 3. The classification laid out by Boyle et al. (2003)
was also considered.
2.3. Injury audit
The Football Association (England) guidelines on injury reporting state that ‘‘an injury can be received during training or
competition and prevents the injured party from participating in
normal training or competition for more than 48 h, not including
the day of the injury’’ (Hawkins, Hulse, Wilkinson, Hodson, &
Gibson, 2001). Absence from training or competition through
sickness and non-musculo-skeletal causes were not included in the
audit.
The nature and mechanism of injury (contact or non-contact),
the number of days training missed, the number of injuries per
season and the number of competitive first team games missed
were recorded by the principal investigator (RC) for the 2007/8
football season (1st July 2007–4th May 2008). An impartial
colleague who was blind to the test protocol collected the injury
9%
30%
Goalkeepers
audit data. Descriptors of the anatomical site of injury, side of injury
(dominant versus non-dominant, left versus right), soft tissue or
osteo/arthrogenic and grade of injury if radiological input had been
sought were utilised. In addition, a diary of the total hours training
and match play was kept throughout the season by the principal
investigator (RC) so that exposure rates could be calculated.
2.4. Statistical analysis
Descriptive statistics using Microsoft Excel and the Vassar
statistical packages (www.faculty.vassar.edu/lowry/VassarStats)
were used to determine mean values pertaining to the demographic data collected. The significance level for all statistical tests
was set at p 0.05, with a power analysis for the study reporting
a value of 69% (www.dssresearch.com) and necessitating the use of
non-parametric statistical analysis. Chi-squared statistical analysis
was used to determine if there was a significant difference in injury
risk and subsequent rehabilitation time between hypermobile and
non-hypermobile participants.
3. Results
Of the thirty-three participants who volunteered for the study,
four had to be followed up by telephone questionnaire to report
injury data as they had been transferred to rival clubs in January
2008. The study group consisted of 51% non-Caucasian and 49%
Caucasian participants.
3.1. Hypermobility status
When analysed by the Stewart and Burden (2004) cut-off
criteria, fourteen participants (42%) scored 4 or more points on the
Beighton scale, qualifying as a hypermobile presentation, whilst
four participants (12%) presented with scores greater than seven,
representing excessive hypermobility. A lower figure of 7 participants (21%) scored 5 or more points to qualify as a hypermobile
presentation as described in the criteria by Boyle et al. (2003). The
mean (1SD) Beighton score for the group was 3.3 (2.8). Fig. 3
represents the spread of scores on the Beighton scale for the group.
When the Beighton results were examined by the number of
positive findings for each of the nine joints and manoeuvres tested,
it was clearly evident that the knee joints presented with the
highest number of positive tests (45 tests/68%).
Defenders
34%
Midfielders
Attackers
27%
Fig. 2. A pie chart representing the positional make up of the football squad for season
2007–2008.
Table 2
The nine point Beighton hypermobility scale (0 ¼ negative test, 1 ¼ positive test).
Test participant ability to:
Left
Right
Passively extend 5th metacarpophalangeal joint >90
Passively appose thumb to volar aspect of forearm
Passively hyperextend elbow >10
Passively hyperextend knee >10
Actively place hands flat on floor without bending knees
Total¼
0/1
0/1
0/1
0/1
0/1
0/1
0/1
0/1
0/1
/9
R. Collinge, J.V. Simmonds / Physical Therapy in Sport 10 (2009) 91–96
Table 3
The subcategories of the Beighton scale (adapted from Stewart & Burden, 2004).
Beighton score
Subject group
0–3
4–6
7–9
Tight/non-hypermobile
Hypermobile
Excessively hypermobile
Days Training
Missed
94
80
71
60
31
40
20
0
Beighton +ve
Beighton -ve
Beighton Classification
3.2. Injury risk
The total number of injuries sustained by the squad was 67 over
the season 2007/8.
Data analysis with a Beighton cut-off score of 4, showed a nonsignificant difference in individual injury risk between the hypermobile (6.2 injuries/1000 h exposure) and non-hypermobile
groups (6.3 injuries/1000 h exposure, c2, p ¼ 0.22).
3.3. Number of training days and competitive games missed
through musculo-skeletal injury
Fig. 4. A bar chart representing the average number of days training missed by
individuals through musculo-skeletal injury when screened using the Beighton scale
(cut-off score set at 4 or more in accordance with Stewart & Burden, 2004).
When the data were divided into osteo/arthrogenic injury as
compared to soft tissue injuries, no significant difference was found
between the two groups (c2, p ¼ 0.92).
4. Discussion
4.1. Hypermobility status
3.4. Nature of injuries sustained
A total of 28 injuries were sustained by the hypermobile group
and 39 in the non-hypermobile group. The nature of injury
between contact and non-contact mechanisms failed to produce
any statistically significant findings between the hypermobile and
non-hypermobile subject groups (c2, p ¼ 0.79).
3.5. Anatomical site of injury
Frequency
The lower limb was the main site of injury across both groups
(83% of injuries). The knee and ankle joints were the most common
sites of injury, with the former being most commonly injured in
the non-hypermobile group (6 injuries as compared to 4 in the
hypermobile group) and the ankle most commonly injured in
the hypermobile participants (6 injuries as compared to 4 in the
non-hypermobile group).
10
9
8
7
6
5
4
3
2
1
0
9
7
4
3
3
3
2
1
1
0
0
1
2
3
4
5
6
7
8
9
Beighton Score
Fig. 3. A bar chart representing the distribution of Beighton scores across the thirtythree professional footballers.
The prevalence of JH noted in the current study at 42% was
higher than previously reported in the literature by Stewart and
Burden (32%, 2004) and Decoster et al. (24%, 1999). These different
values may be attributable to different sized participant groups in
these studies, the different Beighton cut-off score selected in the
latter study or potential differences in ethnic backgrounds and
competition demands between the sporting groups studied. If
a cut-off score of 5 is selected, then the incidence of hypermobility
in the current study is reduced to 21% and may therefore correspond more favourably to the findings of Decoster et al. (1999).
Although the sample size limits the external validity of this
preliminary study, interesting and important trends were seen.
Firstly, it can be seen that there were a high number of positive
Beighton test results for knee hyperextension (68%) in accordance
with the 69% found by Decoster et al. (1999) in male and female
lacrosse players. Ramesh et al. (2005) found a knee hyperextension
prevalence of 37% in a control group that suggests that participants
in kicking sports may well show increased levels of knee hyperextension. Further research is required to clarify whether these
findings are common to the non-sporting population and
comparison to other sports is required. Extrapolation of the
research findings could then be pooled to determine if such
presentations are attributable to genetic factors or are indeed
acquired through participation in football, especially when the
mean (1SD) number of career years in the professional game was
7 years (4.8 years) in this study. Grahame (1999) comments that
joint range can be increased into the hypermobile range by the
sheer hard work of training. It must be considered that with the
development of Football Academies, designed to nurture young
Number Of Games
Missed
The results showed that with the Beighton cut-off score set at 4,
individuals in the hypermobile group missed on average 71 days’
training as compared to 31 days in the non-hypermobile group over
season 2007/8 (c2, p ¼ 0.08, Fig. 4).
In terms of competitive first team games missed, there was no
statistically significant difference between the two groups over
season 2007/8. It is interesting to note that the hypermobile group
missed on average 12 games as compared to 5 in the non-hypermobile group after sustaining musculo-skeletal injury (c2, p ¼ 0.21,
Fig. 5).
15
12
10
5
5
0
Beighton +ve
Beighton -ve
Beighton Classification
Fig. 5. A bar chart representing the average number of competitive first team games
missed by individuals through musculo-skeletal injury when screened using the
Beighton scale (cut-off score set at 4 or more in accordance with Stewart & Burden,
2004).
R. Collinge, J.V. Simmonds / Physical Therapy in Sport 10 (2009) 91–96
talent, some players had been with the present club since the age of
8 years. This could potentially lead to acquired movement patterns
and the knee joints may be susceptible to hyperextension forces
due to the open chain nature of the kicking action. These findings
may partly explain the high incidence of knee osteoarthritis in
retired professional footballers, reported at 49% (Turner, Barlow, &
Heathcote-Elliott, 2000).
When the Beighton scores were analysed by ethnic group, there
was insufficient data to detect a trend. Further large scale investigation is warranted in this area to validate the conflicting findings
of Smith et al. (2005), who reported an absence of ethnic trends
related to hypermobility levels in female junior netball players, and
Russek (1999) who comments upon African individuals in the
general population presenting with increased levels of hypermobility as compared to Caucasian individuals.
4.2. Injury risk
The average individual injury risk over the 2007/8 season was
2.0 per player, which is higher than the 1.3 injuries per player per
season reported by Hawkins et al. (2001).
The interpretation of this study suggests that the injury risk
appears to be very similar between the hypermobile and nonhypermobile players in this professional football group. The
average injury risk in hypermobile participants (6.2 injuries/1000 h
exposure) is almost identical when compared to the non-hypermobile group (6.3 injuries/1000 h exposure, c2 p ¼ 0.22).
These findings concur with Decoster et al. (1999) who reported
an injury risk of 4.7 injuries/1000 h exposure in male lacrosse
players and 2.3 injuries/1000 h exposure in hypermobile subjects
as compared to a statistically insignificant 3.5/1000 h in the nonhypermobile group. The lower injury exposure rates in the lacrosse
study may be explained by the different Beighton cut-off score
employed by Decoster et al. (1999) and the differing demands of
lacrosse as compared to professional football. The authors also fail
to provide details of the training and match-play demands associated with lacrosse which makes direct comparison difficult. Krivickas and Feinberg (1996) also found that there was no difference
in injury risk between hypermobile and non-hypermobile American Collegiate athletes from a diversity of disciplines. There are
currently no other published studies looking at injury risk and
rehabilitation time for hypermobile athletes in professional football
with which to make direct comparison.
The explanation as to why both participant groups presented
with similar injury exposure risks in the current study may be
attributable to the emphasis placed upon preventative medicine in
the sporting domain. The ethos at the football club studied is based
upon pre-habilitation sessions, where squad members are screened
from a biomechanical and orthopaedic perspective in pre-season.
Players subsequently undertake twice weekly preventative exercise
programmes, set by the Physiotherapy and Strength and Conditioning teams, in order to minimise the risk of injury. Although the
small sample size and difficulty in comparing ‘‘like for like’’ injuries
between the two groups hinders the wider extrapolation of the
findings of this study, the hypermobile group may have been
successfully targeted and set appropriate individual programmes
that may have served to reduce their risk of injury.
4.3. Training days and competitive games missed through
musculo-skeletal injury
The current study has found that there is a strong trend between
the two groups when the data is interpreted in terms of the average
number of training days and first team games missed by individuals
in each group through musculo-skeletal injury. The hypermobile
95
group showed a trend towards increased training time and match
play missed once they had sustained an injury. Although not
statistically proven from this study and direct comparison of
injuries was not possible between players, this relationship is
worthy of further investigation as there is a dearth of data published in this field and future findings could be important in
designing rehabilitation programmes and modifying timeframes
for injured hypermobile individuals.
The trends found in this study infer that although the injury
risk/1000 h is similar between hypermobile and non-hypermobile
professional footballers, when hypermobile players sustain a musculo-skeletal injury, such individuals take longer to get back to
competition. This inference concurs with the work of Hardin et al.
(1997) who suggest that return to play timescales maybe longer in
hypermobile individuals, a factor that may have socioeconomic
consequences for the athlete, team and club alike.
Potential explanations for the extended rehabilitation timeframe in hypermobile individuals may revolve around the inability
of hypermobile individuals to lay down mature (type I) collagen
during the healing process. Hormonal factors, such as IGF-1, may
play a part in the control of this process (Denko & Boja, 2001),
possibly due to its effect on metabolism and the healing process
evident after injury. When this theory is combined with the
decrease in proprioceptive acuity described in hypermobile individuals (Hall et al., 1995; Mallik et al., 1994), the potential risk of reinjury may be assumed to be higher in hypermobile individuals if
an awareness of the extended timescales of rehabilitation is not
considered.
Ferrell et al. (2004) have reported that symptoms associated
with hypermobility can be reduced through compliance with
a closed kinetic chain (CKC) exercise programme and it may be
hypothesised that CKC and proprioceptive exercises should be
included in preventative and rehabilitation programmes for
hypermobile individuals. These findings may be important in
managing the symptoms associated with JH and preventing the
longer term sequelae of osteoarthritis in hypermobile individuals,
as reported by Jonsson et al. (1996).
This study certainly suggests that further research is warranted
in the hypermobility field utilising a larger sample size, with
calculations suggesting that 60 participants would raise the power
of the study to greater than 90% with p 0.05 (www.dssresearch.
com). The findings could provide valuable information for physiotherapy and other rehabilitation professionals when considering
the extended timescales of rehabilitation in hypermobile individuals and delaying the onset of loaded and functional exercises so as
to prevent the risk of re-injury and longer term sequelae.
4.4. Limitations and future developments
Future large scale research is warranted and should aim to
further explore the trends identified in this study. This is the first
piece of work presented examining hypermobility in professional
football and should act as a stimulus for multi-centre trials in the
sporting domain.
The limitations of this provisional study are primarily the small
sample size utilised. The study could be improved upon by incorporating assessment of other joints in addition to those tested by
the Beighton scale. Hakim and Grahame (2003) criticise the
Beighton scale for its lack of consideration of joints that fall outside
of the five tested. To augment the accuracy of data collected, future
studies may therefore incorporate the Contompasis score (McNerney & Johnston, 1979) or the 10 point Hospital del Mar scoring
system designed by Bulbena, Duro, and Porta (1992). These
assessment tools may be more specific than the Beighton scale in
a lower limb dominant sport such as football.
96
R. Collinge, J.V. Simmonds / Physical Therapy in Sport 10 (2009) 91–96
Future research may also include the Revised 1998 Brighton
Criteria (Grahame et al., 2000) and the 5 Point Questionnaire
(Hakim & Grahame, 2003). However, because of the inclusion of
historical hypermobility questioning and chronic pain in the
criteria, these tools may not be sensitive enough to identify the
phenotype in sporting populations. Consideration should be given
to other features included in the Brighton criteria; namely skin
laxity, striae and papyraceous scarring which may assist in identifying the phenotype in young, athletic populations.
Consideration must also be given to the fact that the current
study assessed participants using the Beighton scale after the
completion of training sessions. This provisional study provides an
insight into joint hypermobility in an elite football squad that has
not previously been published in the literature. The post-training
assessment period provided the only viable window of opportunity
to assess participants and therefore the impact of ballistic activity
and changes in visco-elastic properties of the soft tissues are
accepted as limitations of this study.
5. Conclusion
This preliminary study investigated the prevalence of hypermobility in a professional football squad and the incidence of injury
over the 2007/8 season. Although the injury rates were similar
between the hypermobile and non-hypermobile individuals,
interesting trends were observed suggestive of extended return to
play timeframes in hypermobile individuals that warrant further
larger scale studies. The findings are potentially important for
rehabilitation professionals in designing and implementing rehabilitation programmes so as to minimise the risk of re-injury and
the socioeconomic implications that this may bring in sporting
populations.
Future studies should look to use the template of this study so
that trends in hypermobility level and return to play timeframes
can be explored further and extrapolated to the Sports Medicine
fraternity.
Conflict of interest
There are no financial or personal conflicts of interest with any
other person or organisation that could inappropriately influence
this work.
Funding
None.
Ethical approval
Ethical approval was obtained from the University College London ethics committee. Informed consent was obtained from all
participants prior to entering the study.
Acknowledgements
University College and King’s College London MSc Sports Physiotherapy Departments and Players and Staff from the Championship Football Club used in this study.
References
Al Rawi, Z., & Nessan, A. H. (1997). Joint hypermobility in patients with chondromalacia patellae. British Journal of Rheumatology, 36, 1324–1327.
Beighton, P. H., Soskolne, L., & Solomon, C. L. (1973). Articular mobility in an African
population. Annals of the Rheumatic Diseases, 32, 413–417.
Boyle, K. L., Witt, P., & Riegger-Krugh, C. (2003). Intrarater and interrater reliability
of the Beighton and Horan Joint Mobility Index. Journal of Athletic Training,
38(4), 281–285.
Bulbena, A., Duro, J. C., & Porta, M. (1992). Clinical assessment of hypermobility of
joints: assembling criteria. Journal of Rheumatology, 19, 115–122.
Dadebo, B., White, J., & George, K. P. (2004). A survey of flexibility training protocols
and hamstring strains in professional football clubs in England. British Journal of
Sports Medicine, 38, 388–394.
Decoster, L. C., Bernier, J. N., Lindsay, R. H., & Vailas, J. C. (1999). Generalised joint
hypermobility and its relationship patterns among NCAA lacrosse players.
Journal of Athletic Training, 34(2), 99–105.
Denko, C. W., & Boja, B. (2001). Growth hormone, insulin and insulin like growth
factor-1 in hypermobility syndrome. Journal of Rheumatology, 28(7), 1666–1669.
Ferrell, W. R., Tennant, N., Sturrock, R. D., Ashton, L., Creed, G., Brydson, G., et al. (2004).
Amelioration of symptoms by enhancement of proprioception in patients with
joint hypermobility syndrome. Arthritis and Rheumatism, 50(10), 3323–3328.
Grahame, R. (1999). Joint hypermobility and genetic collagen disorders: are they
related? Archives of Disease in Childhood, 80, 188–191.
Grahame, R. (2003). Hypermobility and hypermobility syndrome. In R. Keer, &
R. Grahame (Eds.), Hypermobility syndrome, recognition and management for
physiotherapists (pp. 1–14). Edinburgh: Butterworth Heinemann.
Grahame, R., Bird, H. A., & Child, A. (2000). The revised (Brighton 1998) criteria for
the diagnosis of benign joint hypermobility syndrome. Journal of Rheumatology,
27, 1777–1779.
Hakim, A. J., & Grahame, R. (2003). Joint hypermobility syndrome: an update for
clinicians. International Journal of Advances in Rheumatology, 1(4), 131–138.
Hall, M. G., Ferrell, W. R., Sturrock, R. D., Hamblen, D. L., & Baxendale, R. H. (1995).
The effect of the hypermobility syndrome on knee joint proprioception. British
Journal of Rheumatology, 34, 121–125.
Hardin, J. A., Voight, M. L., Blackburn, T. A., Canner, G. C., & Soffer, S. R. (1997). The
effects of ‘‘decelerated’’ rehabilitation following anterior cruciate ligament
reconstruction on a hyperelastic female adolescent: a case study. The Journal of
Orthopaedic and Sports Physical Therapy, 26, 29–34.
Hawkins, R. D., Hulse, M. A., Wilkinson, C., Hodson, A., & Gibson, M. (2001). The
association football medical research programme: an audit of injuries in
professional football. British Journal of Sports Medicine, 35, 43–47.
Jonsson, H., Valtysdottir, S. T., Kjartansson, O., & Brekkan, A. (1996). Hypermobility
associated with osteoarthritis of the thumb base: a clinical and radiological
subset of hand osteoarthritis. Annals of the Rheumatic Diseases, 55, 540–543.
Krivickas, L. S., & Feinberg, J. H. (1996). Lower extremity injuries in college athletes:
relation between ligamentous laxity and lower extremity muscle tightness.
Archives of Physical Medicine and Rehabilitation, 77, 1139–1143.
Malfait, F., Hakim, A. J., De Paepe, A., & Grahame, R. (2006). The genetic basis of the
joint hypermobility syndromes. Rheumatology, 45, 502–507.
Mallik, A. K., Ferrell, W. R., McDonald, A. G., & Sturrock, R. D. (1994). Impaired
proprioceptive acuity at the proximal interphalangeal joint in patients with the
hypermobility syndrome. British Journal of Rheumatology, 33, 631–637.
McNerney, J. E., & Johnston, W. B. (1979). Generalized ligament laxity, hallux
abducto-valgus and the first metatarsocuneiform joint. Journal of the American
Podiatric Medical Association, 69, 69–82.
Middleditch, A. (2003). Management of the hypermobile adolescent. In R. Keer, &
R. Grahame (Eds.), Hypermobility syndrome, recognition and management for
physiotherapists (pp. 51–66). Edinburgh: Butterworth Heinemann.
Myer, G. D., Ford, K. R., Paterno, M. V., Nick, T. G., & Hewett, T. E. (2008). The effects
of generalised joint laxity on risk of anterior cruciate ligament injury in young
female athletes. American Journal of Sports Medicine, 36(6), 1073–1080.
Nicholas, J. A. (1970). Injuries to knee ligaments. JAMA: The Journal of the American
Medical Association, 212(13), 2236–2239.
Ramesh, R., Von Arx, O., Azzopardi, T., & Schranz, P. J. (2005). The risk of anterior
cruciate ligament rupture with generalised joint laxity. The Journal of Bone and
Joint Surgery. British Volume, 87-B, 800–803.
Russek, L. N. (1999). Hypermobility syndrome. Physical Therapy, 79(6), 591–599.
Schalkwijk, J., Zweers, M. C., Steijlen, P. M., Dean, W. B., Taylor, G., van Vlijmen, I. M.,
et al. (2001). A recessive form of the Ehlers–Danlos syndrome caused by
tenascin-X deficiency. New England Journal of Medicine, 345(16), 1167–1175.
Simmonds, J. V., & Keer, R. J. (2007). Hypermobility and the hypermobility
syndrome. Manual Therapy, 12(4), 298–309.
Smith, R., Damodaran, A. K., Swaminathan, S., Campbell, R., & Barnsley, L. (2005).
Hypermobility and sports injuries in junior netball players. British Journal of
Sports Medicine, 39, 628–631.
Stewart, D. R., & Burden, S. B. (2004). Does generalised ligamentous laxity increase
seasonal incidence of injuries in male first division club rugby players? British
Journal of Sports Medicine, 38, 457–460.
Turner, A. P., Barlow, J. H., & Heathcote-Elliott, C. (2000). Long term health impact of
playing professional football in the United Kingdom. British Journal of Sports
Medicine, 34, 332–337.
Zweers, M. C., Dean, W. B., van Kuppevelt, T. H., Bristow, J., & Schalkwijk, J. (2005).
Elastic fibre abnormalities in hypermobility type Ehlers–Danlos syndrome
patients with tenascin-X mutations. Clinical Genetics, 67(4), 330–334.
Websites
www.dssresearch.com.
www.faculty.vassar.edu/lowry/VassarStats.