You are here

Article Text

Article menu

Download PDFPDF

Original research
A 4-year study of hamstring injury outcomes in elite track and field using the British Athletics rehabilitation approach
Free
  1. Noel Pollock1,2,
  2. Shane Kelly2,3,
  3. Justin Lee4,
  4. Ben Stone2,
  5. Michael Giakoumis2,
  6. George Polglass2,
  7. James Brown2,
  8. Ben MacDonald5
  1. 1 Institute of Sport, Exercise and Health, University College London, London, UK
  2. 2 National Performance Institute, British Athletics Science and Medicine Team, Loughborough, UK
  3. 3 Ballet Healthcare, The Royal Ballet, London, UK
  4. 4 Radiology Department, Fortius Clinic, London, UK
  5. 5 Medical Department, British Cycling, Manchester, UK
  1. Correspondence to Dr Noel Pollock, Institute of Sport, Exercise and Health, University College London, London WC1E 6BT, UK; NPollock{at}britishathletics.org.uk

Abstract

Objectives The British Athletics Muscle Injury Classification (BAMIC) correlates with return to play in muscle injury. The aim of this study was to examine hamstring injury diagnoses and outcomes within elite track and field athletes following implementation of the British Athletics hamstring rehabilitation approach.

Methods All hamstring injuries sustained by elite track and field athletes on the British Athletics World Class Programme between December 2015 and November 2019 that underwent an MRI and had British Athletics medical team prescribed rehabilitation were included. Athlete demographics and specific injury details, including mechanism of injury, self-reported gait phase, MRI characteristics and time to return to full training (TRFT) were contemporaneously recorded.

Results 70 hamstring injuries in 46 athletes (24 women and 22 men, 24.6±3.7 years) were included. BAMIC grade and the intratendon c classification correlated with increased TRFT. Mean TRFT was 18.6 days for the entire cohort. Mean TRFT for intratendon classifications was 34±7 days (2c) and 48±17 days (3c). The overall reinjury rate was 2.9% and no reinjuries were sustained in the intratendon classifications. MRI variables of length and cross-sectional (CSA) area of muscle oedema, CSA of tendon injury and loss of tendon tension were associated with TRFT. Longitudinal length of tendon injury, in the intratendon classes, was not associated with TRFT.

Conclusion The application of BAMIC to inform hamstring rehabilitation in British Athletics results in low reinjury rates and favourable TRFT following hamstring injury. The key MRI variables associated with longer recovery are length and CSA of muscle oedema, CSA of tendon injury and loss of tendon tension.

  • hamstring
  • muscle injury
  • exercise rehabilitation
  • rehabilitation
  • elite performance

Data availability statement

Data are available upon reasonable request. Data is available upon reasonable request.

https://doi.org/10.1136/bjsports-2020-103791

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Hamstring injury is common in sports that require sprinting and acceleration.1–5 The rate of hamstring injury is increasing in elite football and baseball.1 3 Reported recurrence rates in prospective follow-up studies range from 14% to 63%.3 6 7 In track and field, hamstring injury is the most prevalent injury in competition and therefore has significant performance and financial implications for athletes, national governing bodies and international federations.8–10

    The British Athletics Muscle Injury Classification (BAMIC) is an MRI classification system with clearly defined, anatomically focused classes based on the site of injury: myofascial (class a), muscle–tendon junction (class b) or intratendon injury (class c) and a numerical grading system (0–4) based on the extent of injury (table 1).11 It is a reliable classification system that is associated with return to play.7 12–18 A retrospective injury review in British Athletics between 2010 and 2014, reported an increase in time to return to full training (TRFT) and reinjury rate in injuries that extended into the hamstring intramuscular tendon (class c).7 An extended TRFT or increased hamstring reinjury rate in injuries that involve the intramuscular tendon has also been seen in other studies19–24 although a prospective study, predominately in footballers, has questioned the clinical relevance of injury to the intramuscular tendon in some athlete groups.23 25 The muscle tendon unit demands in a return to elite sprinting are likely to be significantly increased relative to field sports, such as football, particularly at subelite levels, where maximal speed is less and players may reduce their running speed and distance but still return to play.26 27

    View this table:
    Table 1

    Summary of British Athletics Muscle Injury Classification system

    A small number of hamstring rehabilitation programmes have been published.28–35 As individual muscles have different architecture and function36 37 and there are different prognostic outcomes in certain structural injuries,7 24 38 39 a targeted rehabilitation programme based on the specific injury appears prudent. This is the clinically reasoned, published, rehabilitation approach advocated by the medical team at British Athletics, the national governing body supporting elite British track and field athletes.28 The rehabilitation process is informed by the BAMIC classification and the supporting principle that different tissues within the hamstring undergo different healing processes.40–42

    The aim of this paper is to report hamstring injury diagnoses and outcomes between 2015 and 2019 in an elite track and field programme following the implementation of the BAMIC and the associated targeted rehabilitation approach.

    Methods

    Inclusion criteria

    The British Athletics Medical team provides full time medical care to track and field athletes on the elite Olympic World Class Programme (WCP). Entry to the WCP is by the judgement of a panel of high performance coaches, who deem the athlete to have the potential to win an Olympic or World Championship medal in track and field. All injury episodes were recorded on the Smartabase electronic medical record (EMR). Athletes were eligible for inclusion in this prospective study if they; reported an injury to the posterior thigh during or immediately after a training session or competition between December 2015 and November 2019; the injury was subsequently assessed and the athlete referred for an MRI scan within 7 days of the injury; and rehabilitation was prescribed by the British Athletics medical team.

    Injury characteristics

    The athlete’s gender, age, athletic discipline (Short Sprint, Long Sprint, Other Power, Endurance) and date of injury were recorded. Injury mechanism and other injury specific variables: training venue, mode of onset, self-reported gait phase, coach-reported training intensity and MRI characteristics were also recorded in a bespoke hamstring injury EMR.

    MRI protocol

    Each MRI scan was assessed by the same specialist musculoskeletal radiologist and specific injury characteristics were recorded. The injury was classified using BAMIC (table 1). MRI scans were performed on a 3 T MRI scanner at one of three imaging centres in the UK. All MRI scans included a combination of axial, sagittal and coronal T1-weighted, T2-weighted fat suppressed/proton density-weighted fat suppressed or short T1 inversion recovery (STIR) sequences.43 MRI analysis was performed on McKesson Picture Archive Communication System 64-bit professional workstation. The freehand region of interest tool was used to assess cross-sectional area (CSA) of muscle oedema (high signal change on T2/STIR sequences) and overall muscle area. The same tool was used to assess CSA of tendon abnormal signal versus overall tendon cross section. The length of oedema within the muscle or within the tendon was measured on coronal and/or sagittal images using the distance measuring tool. Loss of tension was defined as lack of taut tendon appearance within the muscle on coronal and/or sagittal images, producing the ‘wavy tendon’ sign.

    Rehabilitation approach

    The rehabilitation programme followed the principles of the British Athletics hamstring injury management approach.28 This is a functional progressive rehabilitation programme that is informed by the nature of the structural injury, the athlete’s functional requirements and a shared decision-making process with the coach and athlete to determine the performance goals. The rehabilitation was provided by full time British Athletics doctors and physiotherapists. If athletes were based abroad, British Athletics’ doctors and physiotherapists would collaborate with a local practitioner to structure the rehabilitation to these principles. The principles of the British Athletics rehabilitation approach are summarised in box 1 and figure 1. While exercise prescription is key, the rehabilitation approach is also characterised by other management principles such as shared decision-making with athlete and coach, risk factor modification and multidiscipline expertise. With respect to exercise prescription, myofascial injuries (class a) are characterised by a functional and expedited return to training, where running based activities are the focus. Muscle–tendon junction injures (class b) are characterised by progressive strengthening and running-based activities guided by the achievement of objective functional milestones. Intratendon injuries (class c) are characterised by a more conservative approach initially, where early lengthening contractions on the tendon are avoided, to respect slower tendon healing.

    Box 1

    Management principles of hamstring injury rehabilitation in British Athletics28

    1. Establish an accurate structural diagnosis and British Athletics Muscle Injury Classification (BAMIC) injury classification

    2. Facilitate the collaborative expertise of the sports science and medicine team

    3. Involve the coach and athlete in shared decision making

    4. Train movements and muscles

    5. Prescribe strength exercises to achieve a specific goal

      • Develop high eccentric force

      • Increase fascicle length to enhance length–tension relationship

      • Develop muscle–tendon unit specificity

      • Develop fatigue resistance

      • Overcome selective muscle inhibition

    6. Apply an individualised non-reductionist model targeting contributing risk factors

    Figure 1
    Figure 1

    Summary of hamstring rehabilitation within British Athletics. Reproduced with permission from MacDonald et al. 28

    Outcome measures

    Injuries were evaluated by gender, location and occurrence (training/competition and self-reported gait phase).

    The TRFT, defined as completing unrestricted sprint efforts at full pace in spikes, was recorded in the athlete’s EMR during the clinical rehabilitation.

    Repeat injury to the same hamstring muscle during rehabilitation (exacerbation) or within 3 months of return to full training (recurrence) was recorded.44 45 A recurrence or exacerbation was recorded if the athlete sustained an acute, sudden return of hamstring pain during exercise, followed by deterioration of functional and clinical tests that required cessation of current activity and subsequent modification of rehabilitation or training for greater than 48 hours.

    Statistical analysis

    The relationships between dependent (TRFT) and explanatory (BAMIC, grade and location of the injury, muscle injured, MRI observations, activity and self reported gait phase at time of injury) variables were tested by Kruskal Wallis, with Dunn’s post-hoc where applicable, or Fisher’s exact test. If athletes suffered a repeat injury during rehabilitation (exacerbation), this injury was excluded from the TRFT analysis but recorded as a reinjury. The relationships between TRFT and the grade and sites (class a–c) of the injuries were examined using linear regression analysis controlled for age and gender. A custom Python script was used to conduct the statistical analysis, with the significance threshold set at p≤0.05.

    Results

    Seventy hamstring injuries in 46 athletes (24 women and 22 men, 24.6±3.7 years) were included in the study. Eighty-seven per cent of injuries occurred in the Sprint/Power group (Short Sprints: 40%, Long Sprints: 25%, Other Power: 21%) and 13% of injuries occurred in the Endurance athlete group. Eighteen athletes sustained more than 1 independent hamstring injury during the study period (15 athletes (two injuries), 2 athletes (3 injuries) and 1 athlete (4 injuries)). In athletes with more than 1 recorded injury during the study, 4 of these were within 1 year and on the same side as the initial injury, but none were at the same location or within the same muscle (grade 2b semitendinosis (ST) followed by a grade 0 injury at 5 months; grade 2a ST followed by grade 0 injury at 4 months, grade 0 followed by grade 1b biceps femoris long head (BFLH) (7 months) and a grade 0 injury followed by a grade 0 injury at 5 months). Only 46% of injuries occurred at the athlete’s usual training venue. Injuries otherwise occurred at different UK training tracks, international training camps or in competition.

    Injury location

    Most injuries occurred in the distal third of the hamstring (43%), with 31% in the proximal third and 26% in the central third. An isolated injury to the BFLH muscle was the most frequently occurring structural muscle injury (70%), followed by injury to the semi-membranosis (19%), multiple muscles (6%), semi-tendinosis (4%) and the short head of biceps femoris (2%). There were 16 grade 0 injuries with a normal MRI scan. The characteristics and location of the injuries are included in tables 2 and 3).

    View this table:
    Table 2

    Characteristics of the study participants and the hamstring injuries

    View this table:
    Table 3

    Location of injury in different British Athletics Muscle Injury Classification (BAMIC) categories

    Injury occurrence

    Seventy-one per cent of injuries occurred during training, 4% during competition warm up and 24% during competition. The coach-reported specific athletic activity undertaken during the injury was sprinting (>90% maximal velocity) in 54% and running at between 50% and 90% maximal velocity in 27%. Injuries occurred during jumping activities in 8% of cases. Athletes self-reported the phase of the running stride at which they felt the injury occur in 89% of injuries. The single most self-reported phase at injury was during the stance phase (37%), other (including unsure, relay change, finishing line) was reported in 26%, terminal swing (18%) and a gradual subacute presentation after the activity in 19% of cases.

    TRFT: BAMIC and TRFT

    The median TRFT for the different BAMIC hamstring injuries is described in table 4 and figure 2. There was a significantly shorter TRFT for grade 0 injuries with respect to every other grade (grade 0 vs grade 1, p=0.03; grade 0 vs grade 2 p<0.001; grade 0 vs grade 3 p<0.001). There was also a significant difference between grade 1 vs grade 2 injuries (p=0.03) and grade 1 and grade 3 (p=0.004). There was no difference in TRFT between grade 2 and grade 3 injuries (p=0.14). There was a highly significant difference in TRFT between class c (intratendon) injuries versus class a (myofascial, p=0.0009) and class c versus class b (muscular or muscle–tendon junction, p=0.0009) injuries. There was no difference in TRFT between classes a and b injuries (p=0.32).

    View this table:
    Table 4

    Time to return to full training in the different British Athletics Muscle Injury Classification (BAMIC) classes

    Figure 2
    Figure 2

    Median±IQR (days) time to return to full training (TRFT) in different British Athletics Muscle Injury Classification (BAMIC) categories.

    MRI variables and TRFT

    In the structural (MRI positive) injuries, the length of muscular oedema (p=0.0001) and CSA of muscular oedema (p=0.0005) were significantly associated with TRFT. The length of intratendon injury was not associated with TRFT (p=0.16) but the CSA of tendon injury was significantly associated with TRFT (p=0.003). Loss of tendon tension was associated with a longer TRFT (p=0.02). There was no significant difference in outcome if the injury site was the distal T junction of the long head and short head of biceps femoris (p=0.85). There was no difference in TRFT between proximal and distal injuries.

    Activity and self-reported gait phase at time of injury and TRFT

    Injuries that were sustained in competition took on average 28.7 days (±18.6) to return to full training compared with 17.7 days (±10.3) for those injuries sustained in training (p=0.09). Intratendon class c injuries occurred more frequently in competition than training (p=0.004). There was no significant difference in outcomes related to the self-reported phase of the running cycle at which the injury occurred or the training activity that was being undertaken at the time of injury.

    Reinjuries

    Only two repeat injuries (recurrence or exacerbation) were sustained in this cohort of 70 hamstring injuries over a 4-year period: in one index 1a and one index 2b injury. This gives a reinjury rate of 2.9% in this elite athlete group.

    Discussion

    The main finding of this paper is that implementing the BAMIC rehabilitation approach to hamstring injuries was associated with a very low reinjury rate in this elite athlete cohort. A total reinjury rate of 2.9%, and no reinjuries within the intratendon classification, compares favourably to previous work in elite track and field athletes and other elite sports where reinjury rates are usually between 12% and 25%, and at times higher in injuries that involve the tendon.7 24 25 46–48 Similarly low reinjury rates have been reported.34 49 Askling et al compared a lengthening rehabilitation protocol to what was described as a conventional rehabilitation protocol.49 Two reinjuries (n=56) were reported in the conventional protocol. However, the average mean time to return of 49 days (lengthening protocol) and 86 days (conventional protocol) was far greater than the mean return times in our study. Mendiguchia et al also reported a 4% reinjury rate using their multifactorial criteria-based progressive algorithm for hamstring rehabilitation, although the study population was limited to footballers with grade 1 muscle injuries.34

    The BAMIC injury grade and the intratendon class c classification are associated with longer TRFT in elite track and field athletes when this rehabilitation approach is prescribed. This study supports previous work that muscle injuries involving the tendon have a longer TRFT but can have an excellent outcome without surgical intervention.7 19–22 24 25 While performance and rehabilitation relevance may vary between different sports, tendon tissue heals more slowly and differently to muscle, requiring collagen synthesis and remodelling within an extracellular matrix scaffold.40 50 The tendon has an important role in muscle–tendon unit force production and demand increases non-linearly as running speed increases.27 The British Athletics rehabilitation approach for tendon injuries is detailed in previous work.28 The loading principles are: an initial longer period of protection for the healing tissue by avoiding eccentric exercises or activities with high ‘elongation stress’51; early isometric loading to facilitate tendon adaptation, progressing to eccentric loads at gradually increasing lengths; and tendon specific loading considerations, including heavy and long-term loading. It is also important to note that the rehabilitation approach is not merely an exercise protocol. A collaborative, informed multidisciplinary team and shared decision-making with athlete and coach are critical principles within the British Athletics rehabilitation approach which enhance communication and sharing of knowledge.28

    This overall rehabilitation approach has not only resulted in very low recurrence rates but also favourable TRFT times (figure 2). The TRFT is lower in all but one injury category (the 2c category), relative to previous work in track and field. In a previous 4-year study from 2010 to 2014, the average TRFT for 1a injuries was 18±4 days; for 1b 18±11 days; for 2b 21±10 days; for 2c 27±7 days and for 3c injuries 84±49 days.7 In this current study, the mean TRFT for the 2c category was longer by 7 days but all other categories had lower TRFT. Of particular note was the 3c class (figure 3) which had a mean TRFT of 48±17 days. These favourable TRFT times suggest that the low recurrence rates in the current study are not a result of an overall slower, cautious rehabilitation approach.

    Figure 3
    Figure 3

    Coronal fat-saturated proton density weighted MRI of a male athlete with acute right mid-thigh hamstring injury. Typical feathery intramuscular oedema is seen surrounding an acute proximal third long head biceps femoris (LHBF) intramuscular tendon injury with loss of tension and wavy tendon sign on the distal side of the injury epicentre. The loss of tension is indicative of a British Athletics Muscle Injury Classification 3c injury. The injury has occurred below the division of the proximal conjoint long head biceps femoris and semitendinosis (ST) tendon.

    The BAMIC grade also correlated with TRFT with this rehabilitation approach. Correlation between BAMIC and RTP has also been found in a number of previous studies.7 13 15–17 Consistent with previous research, grade 0 injuries have a short TRFT with low recurrence. Several radiological features, including length and CSA of muscle oedema, were associated with differences in TRFT. This suggests that these original criteria for BAMIC radiological grading were appropriate and do not need revision. However, only the CSA of tendon involvement and not the longitudinal length of tendon involvement were significantly associated with TRFT. This may be due to tendon tension and muscle–tendon unit integrity being maintained with intratendon longitudinal splits but disrupted with larger CSA tendon injuries. Long intratendon splits are often detected easily in multiple axial MRI sections (figure 4) and the intratendon involvement may extend for many centimetres. An update of the BAMIC system could consider removing or changing the length of tendon injury as a differentiator between classes 2c and 3c. The CSA of tendon injury and loss of tendon tension (waviness) appear to be key prognostic radiological findings in the intratendon class of injury. Consistent with previous work, 1a injuries appear to have a positive prognostic outcome. Further work is required to understand whether myofascial injuries warrant an independent classification and subsequent rehabilitation approach.

    Figure 4
    Figure 4

    Axial fat-saturated proton density weighted MRI of a female athlete with acute proximal third right hamstring injury. There is a coronal-orientated split within the proximal conjoint long head biceps femoris (LHBF) and semitendinosis (ST) tendon (angled arrow). The split involved less than 50% of tendon cross section and extended over a length of 2 cm without loss of tendon tension. Therefore, the injury is classified as 2c by British Athletics Muscle Injury Classification criteria. Note the close relationships of the conjoint and semimembranosus tendons with the sciatic nerve (F=femoral shaft).

    All cases in this prospective study were managed with expert rehabilitation and without surgical intervention or platelet rich plasma (PRP) injection. The low reinjury rates and reasonable TRFT times suggest that non-surgical management is usually appropriate for 2c and 3c injuries to enable return to elite level sprinting. Further work to determine surgical criteria for high grade (particularly 4c) intratendon hamstring injuries is required.

    The most common mechanism of injury reported was high speed sprinting. Unsurprisingly, the most commonly injured muscle was the long head of biceps, which previous research has determined undergoes the largest strain during maximal sprinting.24 37 52–55 One method of describing hamstring injury mechanism is to classify into stretch and sprinting types.21 Stretch mechanism injuries have been associated with more proximal injury to the semimembranosus, while sprinting mechanism is distributed between proximal and distal portions of the long head of bicep femoris.56 Our data are consistent with this, as 43% of injuries occurred in the distal third of long head of bicep compared with 31% in the proximal third.

    A potential explanation for the increased incidence of distal bicep injuries compared with other research49 could be the interaction between architectural differences along the length of the hamstring muscle and aberrant sprinting biomechanics. The distal BFLH has shorter fascicles than the proximal BLFH (6.35 cm vs 7.12 cm) and the ST (6.35 cm vs 15.49 cm).57 This makes it less well adapted for increased excursion of the tendon due to knee extension in the end swing phase.58 59 Excessive knee extension, and trunk flexion, at end swing phase has been suggested as a potential biomechanical fault leading to hamstring injury, as it leads to excessive strain on the bicep.59–61 If the BFLH is less well adapted to accommodate increased knee extension, this will be magnified during faulty running mechanics—such as over-striding—that causes a further increase in knee extension and consequently biomechanical load on the bicep.

    Consistent with previous research in track and field most injuries occurred during training,62 possibly due to greater exposure to high speed running and fatigue in training relative to competition. However, class c injuries occurred relatively more often during competition. As running velocity increases, tendon strain and work done increase in a non-linear manner.27 63 An increased exposure to maximal velocity running in competition may account for the increased proportion of tendon injuries during competition.64

    Strengths and limitations

    The strength of this study was its prospective nature and the creation of a bespoke EMR to assist with consistent radiological and clinical record keeping. The BAMIC rehabilitation approach standardised the principles of rehabilitation while enabling an individualised rehabilitation programme, informed by previous work in this elite track and field group. That the medical staff were working full time within the sport, reporting within a performance team, gives confidence to the recording of the TRFT and identification of reinjury. However, as the BAMIC rehabilitation is intrinsically linked to the structural injury there is the undoubted limitation of a rehabilitation team that was not blinded to the MRI results resulting in risk of bias. This is particularly the case in the intratendon injuries for which the BAMIC approach advocates a 2-week delay in the progression of eccentric type exercise and accelerated running velocity to avoid placing elastic strain on the healing tendon. Thus, we cannot exclude that the MRI variables associated with longer TRFT are biased by the BAMIC rehabilitation protocol. This is mitigated somewhat as the performance culture within British Athletics demands appropriately expedited return to training and the coach and athlete were fully involved in shared decision-making regarding rehabilitation and performance targets. The TRFT relative to the existing literature suggests that rehabilitation outcomes were not inappropriately delayed due to cautious rehabilitation or prescribing rehabilitation progression based on imaging findings and TRFT from previously published studies.

    Conclusion

    This study demonstrates the successful application of the British Athletics rehabilitation approach following BAMIC hamstring classification. The particularly low reinjury rates suggest that this targeted rehabilitation approach is helpful in successfully returning intratendon hamstring injuries to elite track and field. The key MRI variables are length and cross-section of muscle oedema, CSA of tendon involvement and loss of tension.

    What are the findings?

    • British Athletics Muscle Injury Classification injury grade and intratendon class c injury are associated with increased time to return to full training in elite track and field.

    • The cross-sectional area of tendon involvement and loss of tension are associated with time to return to full training rather than length of tendon involvement.

    How might it impact on clinical practice in the future?

    • The application of the British Athletics hamstring rehabilitation principles results in favourable time to return to full training with low reinjury rates.

    • Hamstring injuries that involve the intramuscular tendon can be managed successfully with rehabilitation rather than surgery.

    Data availability statement

    Data are available upon reasonable request. Data is available upon reasonable request.

    Ethics statements

    Patient consent for publication

    Ethics approval

    Ethical approval was granted by University College London ethics panel.

    Acknowledgments

    The authors would like to thank Julie Hayton and Sammy Thorsen for their assistance with data collection and the development of the bespoke hamstring electronic medical record. We acknowledge Dr Robin Chakraverty's design of the infographic that is reproduced here in Figure 1. We would also like to thank all members of the British Athletics medical team, past and present, who have helped with the development and implementation of BAMIC and our muscle injury strategy. The authors also acknowledge the support of Sybermedica, who provided complimentary access to their PACSMail image transfer network for this study.

    References

    1. ↵
      2. Ekstrand J ,
      3. Waldén M ,
      4. Hägglund M
      . Hamstring injuries have increased by 4% annually in men's professional football, since 2001: a 13-year longitudinal analysis of the UEFA Elite Club injury study. Br J Sports Med 2016;50:731–7.doi:10.1136/bjsports-2015-095359 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26746908
      OpenUrlAbstract/FREE Full Text
    2. ↵
      2. Jones A ,
      3. Jones G ,
      4. Greig N , et al
      . Epidemiology of injury in English professional football players: a cohort study. Phys Ther Sport 2019;35:18–22.doi:10.1016/j.ptsp.2018.10.011 pmid:http://www.ncbi.nlm.nih.gov/pubmed/30408703
      OpenUrlPubMed
    3. ↵
      2. Okoroha KR ,
      3. Conte S ,
      4. Makhni EC , et al
      . Hamstring injury trends in major and minor League baseball: epidemiological findings from the major league baseball health and injury tracking system. Orthop J Sports Med 2019;7:232596711986106.doi:10.1177/2325967119861064 pmid:http://www.ncbi.nlm.nih.gov/pubmed/31431899
      OpenUrlPubMed
    4. ↵
      2. Ahmad CS ,
      3. Dick RW ,
      4. Snell E , et al
      . Major and minor League baseball hamstring injuries: epidemiologic findings from the major league baseball injury surveillance system. Am J Sports Med 2014;42:1464–70.doi:10.1177/0363546514529083 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24727933
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      2. Brooks JHM ,
      3. Fuller CW ,
      4. Kemp SPT , et al
      . Incidence, risk, and prevention of hamstring muscle injuries in professional rugby Union. Am J Sports Med 2006;34:1297–306.doi:10.1177/0363546505286022 pmid:http://www.ncbi.nlm.nih.gov/pubmed/16493170
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      2. de Visser HM ,
      3. Reijman M ,
      4. Heijboer MP , et al
      . Risk factors of recurrent hamstring injuries: a systematic review. Br J Sports Med 2012;46:124–30.doi:10.1136/bjsports-2011-090317 pmid:http://www.ncbi.nlm.nih.gov/pubmed/22011915
      OpenUrlAbstract/FREE Full Text
    7. ↵
      2. Pollock N ,
      3. Patel A ,
      4. Chakraverty J , et al
      . Time to return to full training is delayed and recurrence rate is higher in intratendinous ('c') acute hamstring injury in elite track and field athletes: clinical application of the British Athletics Muscle Injury Classification. Br J Sports Med 2016;50:305–10.doi:10.1136/bjsports-2015-094657 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26888072
      OpenUrlAbstract/FREE Full Text
    8. ↵
      2. Edouard P ,
      3. Branco P ,
      4. Alonso J-M
      . Muscle injury is the principal injury type and hamstring muscle injury is the first injury diagnosis during top-level international athletics championships between 2007 and 2015. Br J Sports Med 2016;50:619–30.doi:10.1136/bjsports-2015-095559 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26887415
      OpenUrlAbstract/FREE Full Text
    9. ↵
      2. Alonso J-M ,
      3. Edouard P ,
      4. Fischetto G , et al
      . Determination of future prevention strategies in elite track and field: analysis of Daegu 2011 IAAF Championships injuries and illnesses surveillance. Br J Sports Med 2012;46:505–14.doi:10.1136/bjsports-2012-091008 pmid:http://www.ncbi.nlm.nih.gov/pubmed/22522588
      OpenUrlAbstract/FREE Full Text
    10. ↵
      2. Raysmith BP ,
      3. Drew MK
      . Performance success or failure is influenced by weeks lost to injury and illness in elite Australian track and field athletes: a 5-year prospective study. J Sci Med Sport 2016;19:778–83.doi:10.1016/j.jsams.2015.12.515 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26839047
      OpenUrlPubMed
    11. ↵
      2. Pollock N ,
      3. James SLJ ,
      4. Lee JC , et al
      . British athletics muscle injury classification: a new grading system. Br J Sports Med 2014;48:1347–51.doi:10.1136/bjsports-2013-093302 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25031367
      OpenUrlAbstract/FREE Full Text
    12. ↵
      2. Patel A ,
      3. Chakraverty J ,
      4. Pollock N , et al
      . British athletics muscle injury classification: a reliability study for a new grading system. Clin Radiol 2015;70:1414–20.doi:10.1016/j.crad.2015.08.009 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26385202
      OpenUrlPubMed
    13. ↵
      2. Wangensteen A ,
      3. Guermazi A ,
      4. Tol JL , et al
      . New MRI muscle classification systems and associations with return to sport after acute hamstring injuries: a prospective study. Eur Radiol 2018;28:3532–41.doi:10.1007/s00330-017-5125-0 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29460072
      OpenUrlPubMed
    14. ↵
      2. Wangensteen A ,
      3. Tol JL ,
      4. Roemer FW , et al
      . Intra- and interrater reliability of three different MRI grading and classification systems after acute hamstring injuries. Eur J Radiol 2017;89:182–90.doi:10.1016/j.ejrad.2017.02.010 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28267537
      OpenUrlPubMed
    15. ↵
      2. Pezzotta G ,
      3. Pecorelli A ,
      4. Querques G , et al
      . Mri characteristics of adductor longus lesions in professional football players and prognostic factors for return to play. Eur J Radiol 2018;108:52–8.doi:10.1016/j.ejrad.2018.09.018 pmid:http://www.ncbi.nlm.nih.gov/pubmed/30396671
      OpenUrlPubMed
    16. ↵
      2. Pezzotta G ,
      3. Querques G ,
      4. Pecorelli A , et al
      . Mri detection of soleus muscle injuries in professional football players. Skeletal Radiol 2017;46:1513–20.doi:10.1007/s00256-017-2729-z pmid:http://www.ncbi.nlm.nih.gov/pubmed/28770309
      OpenUrlPubMed
    17. ↵
      2. Biglands JD ,
      3. Grainger AJ ,
      4. Robinson P , et al
      . Mri in acute muscle tears in athletes: can quantitative T2 and DTI predict return to play better than visual assessment? Eur Radiol 2020;30:6603–13.doi:10.1007/s00330-020-06999-z pmid:http://www.ncbi.nlm.nih.gov/pubmed/32666321
      OpenUrlPubMed
    18. ↵
      2. Nescolarde L ,
      3. Terricabras J ,
      4. Mechó S , et al
      . Differentiation between tendinous, myotendinous and myofascial injuries by L-BIA in professional football players. Front Physiol 2020;11:574124. doi:10.3389/fphys.2020.574124 pmid:http://www.ncbi.nlm.nih.gov/pubmed/33013488
      OpenUrlPubMed
    19. ↵
      2. Comin J ,
      3. Malliaras P ,
      4. Baquie P , et al
      . Return to competitive play after hamstring injuries involving disruption of the central tendon. Am J Sports Med 2013;41:111–5.doi:10.1177/0363546512463679 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23111807
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      2. Brukner P ,
      3. Connell D
      . 'Serious thigh muscle strains': beware the intramuscular tendon which plays an important role in difficult hamstring and quadriceps muscle strains. Br J Sports Med 2016;50:205–8.doi:10.1136/bjsports-2015-095136 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26519522
      OpenUrlAbstract/FREE Full Text
    21. ↵
      2. Askling CM ,
      3. Tengvar M ,
      4. Saartok T , et al
      . Acute first-time hamstring strains during high-speed running: a longitudinal study including clinical and magnetic resonance imaging findings. Am J Sports Med 2007;35:197–206.doi:10.1177/0363546506294679 pmid:http://www.ncbi.nlm.nih.gov/pubmed/17170160
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      2. Cohen SB ,
      3. Towers JD ,
      4. Zoga A , et al
      . Hamstring injuries in professional football players: magnetic resonance imaging correlation with return to play. Sports Health 2011;3:423–30.doi:10.1177/1941738111403107 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23016038
      OpenUrlCrossRefPubMed
    23. ↵
      2. van der Made AD ,
      3. Almusa E ,
      4. Whiteley R , et al
      . Intramuscular tendon involvement on MRI has limited value for predicting time to return to play following acute hamstring injury. Br J Sports Med 2018;52:83–8.doi:10.1136/bjsports-2017-097659 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28903949
      OpenUrlAbstract/FREE Full Text
    24. ↵
      2. Eggleston L ,
      3. McMeniman M ,
      4. Engstrom C
      . High-Grade intramuscular tendon disruption in acute hamstring injury and return to play in Australian football players. Scand J Med Sci Sports 2020;30:1073–82.doi:10.1111/sms.13642 pmid:http://www.ncbi.nlm.nih.gov/pubmed/32096248
      OpenUrlPubMed
    25. ↵
      2. van der Made AD ,
      3. Almusa E ,
      4. Reurink G , et al
      . Intramuscular tendon injury is not associated with an increased hamstring reinjury rate within 12 months after return to play. Br J Sports Med 2018;52:1261–6.doi:10.1136/bjsports-2017-098725 pmid:http://www.ncbi.nlm.nih.gov/pubmed/29654058
      OpenUrlAbstract/FREE Full Text
    26. ↵
      2. Whiteley R ,
      3. Massey A ,
      4. Gabbett T , et al
      . Match high-speed running distances are often suppressed after return from hamstring strain injury in professional footballers. Sports Health 2020:194173812096445. doi:10.1177/1941738120964456 pmid:http://www.ncbi.nlm.nih.gov/pubmed/33151808
    27. ↵
      2. Schache AG ,
      3. Dorn TW ,
      4. Williams GP , et al
      . Lower-Limb muscular strategies for increasing running speed. J Orthop Sports Phys Ther 2014;44:813–24.doi:10.2519/jospt.2014.5433 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25103134
      OpenUrlCrossRefPubMed
    28. ↵
      2. Macdonald B ,
      3. McAleer S ,
      4. Kelly S , et al
      . Hamstring rehabilitation in elite track and field athletes: applying the British athletics muscle injury classification in clinical practice. Br J Sports Med 2019;53:1464–73.doi:10.1136/bjsports-2017-098971 pmid:http://www.ncbi.nlm.nih.gov/pubmed/31300391
      OpenUrlAbstract/FREE Full Text
    29. ↵
      2. Mendiguchia J ,
      3. Brughelli M
      . A return-to-sport algorithm for acute hamstring injuries. Phys Ther Sport 2011;12:2–14.doi:10.1016/j.ptsp.2010.07.003 pmid:http://www.ncbi.nlm.nih.gov/pubmed/21256444
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      2. Heiderscheit BC ,
      3. Sherry MA ,
      4. Silder A , et al
      . Hamstring strain injuries: recommendations for diagnosis, rehabilitation, and injury prevention. J Orthop Sports Phys Ther 2010;40:67–81.doi:10.2519/jospt.2010.3047 pmid:http://www.ncbi.nlm.nih.gov/pubmed/20118524
      OpenUrlCrossRefPubMedWeb of Science
    31. ↵
      2. Askling CM ,
      3. Tengvar M ,
      4. Thorstensson A
      . Acute hamstring injuries in Swedish elite football: a prospective randomised controlled clinical trial comparing two rehabilitation protocols. Br J Sports Med 2013;47:953–9.doi:10.1136/bjsports-2013-092165 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23536466
      OpenUrlAbstract/FREE Full Text
    32. ↵
      2. Sherry MA ,
      3. Johnston TS ,
      4. Heiderscheit BC
      . Rehabilitation of acute hamstring strain injuries. Clin Sports Med 2015;34:263–84.doi:10.1016/j.csm.2014.12.009 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25818713
      OpenUrlPubMed
    33. ↵
      2. Silder A ,
      3. Sherry MA ,
      4. Sanfilippo J , et al
      . Clinical and morphological changes following 2 rehabilitation programs for acute hamstring strain injuries: a randomized clinical trial. J Orthop Sports Phys Ther 2013;43:284–99.doi:10.2519/jospt.2013.4452 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23485730
      OpenUrlCrossRefPubMed
    34. ↵
      2. Mendiguchia J ,
      3. Martinez-Ruiz E ,
      4. Edouard P , et al
      . A multifactorial, Criteria-based progressive algorithm for hamstring injury treatment. Med Sci Sports Exerc 2017;49:1482–92.doi:10.1249/MSS.0000000000001241 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28277402
      OpenUrlPubMed
    35. ↵
      2. Sherry MA ,
      3. Best TM
      . A comparison of 2 rehabilitation programs in the treatment of acute hamstring strains. J Orthop Sports Phys Ther 2004;34:116–25.doi:10.2519/jospt.2004.34.3.116 pmid:http://www.ncbi.nlm.nih.gov/pubmed/15089024
      OpenUrlCrossRefPubMedWeb of Science
    36. ↵
      2. Lieber RL ,
      3. Ward SR
      . Skeletal muscle design to meet functional demands. Philos Trans R Soc Lond B Biol Sci 2011;366:1466–76.doi:10.1098/rstb.2010.0316 pmid:http://www.ncbi.nlm.nih.gov/pubmed/21502118
      OpenUrlCrossRefPubMed
    37. ↵
      2. Schache AG ,
      3. Dorn TW ,
      4. Blanch PD , et al
      . Mechanics of the human hamstring muscles during sprinting. Med Sci Sports Exerc 2012;44:647–58.doi:10.1249/MSS.0b013e318236a3d2 pmid:http://www.ncbi.nlm.nih.gov/pubmed/21912301
      OpenUrlCrossRefPubMedWeb of Science
    38. ↵
      2. Pedret C ,
      3. Balius R ,
      4. Blasi M , et al
      . Ultrasound classification of medial gastrocnemious injuries. Scand J Med Sci Sports 2020;30:2456–65.doi:10.1111/sms.13812 pmid:http://www.ncbi.nlm.nih.gov/pubmed/32854168
      OpenUrlPubMed
    39. ↵
      2. Pedret C ,
      3. Rodas G ,
      4. Balius R , et al
      . Return to play after soleus muscle injuries. Orthop J Sports Med 2015;3:232596711559580.doi:10.1177/2325967115595802 pmid:http://www.ncbi.nlm.nih.gov/pubmed/26674181
      OpenUrlPubMed
    40. ↵
      2. James R ,
      3. Kesturu G ,
      4. Balian G , et al
      . Tendon: biology, biomechanics, repair, growth factors, and evolving treatment options. J Hand Surg Am 2008;33:102–12.doi:10.1016/j.jhsa.2007.09.007 pmid:http://www.ncbi.nlm.nih.gov/pubmed/18261674
      OpenUrlCrossRefPubMed
    41. ↵
      2. Järvinen TAH ,
      3. Järvinen TLN ,
      4. Kääriäinen M , et al
      . Muscle injuries: biology and treatment. Am J Sports Med 2005;33:745–64.doi:10.1177/0363546505274714 pmid:http://www.ncbi.nlm.nih.gov/pubmed/15851777
      OpenUrlCrossRefPubMedWeb of Science
    42. ↵
      2. Zügel M ,
      3. Maganaris CN ,
      4. Wilke J , et al
      . Fascial tissue research in sports medicine: from molecules to tissue adaptation, injury and diagnostics: consensus statement. Br J Sports Med 2018;52:1497. doi:10.1136/bjsports-2018-099308 pmid:http://www.ncbi.nlm.nih.gov/pubmed/30072398
      OpenUrlAbstract/FREE Full Text
    43. ↵
      2. Patel A ,
      3. Chakraverty J ,
      4. Pollock N , et al
      . British athletics muscle injury classification: a validity study for a new grading system. Press, 2015.
    44. ↵
      2. Bahr R ,
      3. Clarsen B ,
      4. Derman W , et al
      . International Olympic Committee consensus statement: methods for recording and reporting of epidemiological data on injury and illness in sport 2020 (including STROBE extension for sport injury and illness surveillance (STROBE-SIIS)). Br J Sports Med 2020;54:372–89.doi:10.1136/bjsports-2019-101969 pmid:http://www.ncbi.nlm.nih.gov/pubmed/32071062
      OpenUrlAbstract/FREE Full Text
    45. ↵
      2. Timpka T ,
      3. Alonso J-M ,
      4. Jacobsson J , et al
      . Injury and illness definitions and data collection procedures for use in epidemiological studies in athletics (track and field): consensus statement. Br J Sports Med 2014;48:483–90.doi:10.1136/bjsports-2013-093241 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24620036
      OpenUrlAbstract/FREE Full Text
    46. ↵
      2. De Vos R-J ,
      3. Reurink G ,
      4. Goudswaard G-J , et al
      . Clinical findings just after return to play predict hamstring re-injury, but baseline MRI findings do not. Br J Sports Med 2014;48:1377–84.doi:10.1136/bjsports-2014-093737 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25037201
      OpenUrlAbstract/FREE Full Text
    47. ↵
      2. Tol JL ,
      3. Hamilton B ,
      4. Eirale C , et al
      . At return to play following hamstring injury the majority of professional football players have residual isokinetic deficits. Br J Sports Med 2014;48:1364–9.doi:10.1136/bjsports-2013-093016 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24493666
      OpenUrlAbstract/FREE Full Text
    48. ↵
      2. Hägglund M ,
      3. Waldén M ,
      4. Ekstrand J
      . Injury recurrence is lower at the highest professional football level than at national and amateur levels: does sports medicine and sports physiotherapy deliver? Br J Sports Med 2016;50:751–8.doi:10.1136/bjsports-2015-095951 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27015858
      OpenUrlAbstract/FREE Full Text
    49. ↵
      2. Askling CM ,
      3. Tengvar M ,
      4. Tarassova O , et al
      . Acute hamstring injuries in Swedish elite sprinters and jumpers: a prospective randomised controlled clinical trial comparing two rehabilitation protocols. Br J Sports Med 2014;48:532–9.doi:10.1136/bjsports-2013-093214 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24620041
      OpenUrlAbstract/FREE Full Text
    50. ↵
      2. Liu SH ,
      3. Yang RS ,
      4. al-Shaikh R , et al
      . Collagen in tendon, ligament, and bone healing. A current review. Clin Orthop Relat Res 1995;318:265–78.pmid:http://www.ncbi.nlm.nih.gov/pubmed/7671527
      OpenUrlPubMed
    51. ↵
      2. Guex K ,
      3. Millet GP
      . Conceptual framework for strengthening exercises to prevent hamstring strains. Sports Med 2013;43:1207–15.doi:10.1007/s40279-013-0097-y pmid:http://www.ncbi.nlm.nih.gov/pubmed/24062275
      OpenUrlCrossRefPubMed
    52. ↵
      2. Crema MD ,
      3. Guermazi A ,
      4. Tol JL , et al
      . Acute hamstring injury in football players: association between anatomical location and extent of injury-A large single-center MRI report. J Sci Med Sport 2016;19:317–22.doi:10.1016/j.jsams.2015.04.005 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25921622
      OpenUrlCrossRefPubMed
    53. ↵
      2. Ekstrand J ,
      3. Lee JC ,
      4. Healy JC
      . Mri findings and return to play in football: a prospective analysis of 255 hamstring injuries in the UEFA elite Club injury study. Br J Sports Med 2016;50:738–43.doi:10.1136/bjsports-2016-095974 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27084882
      OpenUrlAbstract/FREE Full Text
    54. ↵
      2. Higashihara A ,
      3. Ono T ,
      4. Kubota J , et al
      . Functional differences in the activity of the hamstring muscles with increasing running speed. J Sports Sci 2010;28:1085–92.doi:10.1080/02640414.2010.494308 pmid:http://www.ncbi.nlm.nih.gov/pubmed/20672221
      OpenUrlCrossRefPubMedWeb of Science
    55. ↵
      2. Higashihara A ,
      3. Nagano Y ,
      4. Ono T , et al
      . Differences in hamstring activation characteristics between the acceleration and maximum-speed phases of sprinting. J Sports Sci 2018;36:1–6.doi:10.1080/02640414.2017.1375548 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28873030
      OpenUrlPubMed
    56. ↵
      2. Askling CM ,
      3. Tengvar M ,
      4. Saartok T , et al
      . Acute first-time hamstring strains during slow-speed stretching: clinical, magnetic resonance imaging, and recovery characteristics. Am J Sports Med 2007;35:1716–24.doi:10.1177/0363546507303563 pmid:http://www.ncbi.nlm.nih.gov/pubmed/17567821
      OpenUrlCrossRefPubMedWeb of Science
    57. ↵
      2. Kellis E ,
      3. Galanis N ,
      4. Natsis K , et al
      . Muscle architecture variations along the human semitendinosus and biceps femoris (long head) length. J Electromyogr Kinesiol 2010;20:1237–43.doi:10.1016/j.jelekin.2010.07.012 pmid:http://www.ncbi.nlm.nih.gov/pubmed/20727788
      OpenUrlCrossRefPubMed
    58. ↵
      2. Kellis E ,
      3. Intra- KE
      . Intra- and Inter-Muscular Variations in Hamstring Architecture and Mechanics and Their Implications for Injury: A Narrative Review. Sports Med 2018;48:2271–83.doi:10.1007/s40279-018-0975-4 pmid:http://www.ncbi.nlm.nih.gov/pubmed/30117053
      OpenUrlPubMed
    59. ↵
      2. Kenneally-Dabrowski C ,
      3. Brown NAT ,
      4. Warmenhoven J , et al
      . Late swing running mechanics influence hamstring injury susceptibility in elite rugby athletes: a prospective exploratory analysis. J Biomech 2019;92:112–9.doi:10.1016/j.jbiomech.2019.05.037 pmid:http://www.ncbi.nlm.nih.gov/pubmed/31176462
      OpenUrlPubMed
    60. ↵
      2. Thelen DG ,
      3. Chumanov ES ,
      4. Best TM , et al
      . Simulation of biceps femoris musculotendon mechanics during the swing phase of sprinting. Med Sci Sports Exerc 2005;37:1931–8.doi:10.1249/01.mss.0000176674.42929.de pmid:http://www.ncbi.nlm.nih.gov/pubmed/16286864
      OpenUrlCrossRefPubMedWeb of Science
    61. ↵
      2. Pinniger GJ ,
      3. Steele JR ,
      4. Groeller H
      . Does fatigue induced by repeated dynamic efforts affect hamstring muscle function? Med Sci Sports Exerc 2000;32:647–53.doi:10.1097/00005768-200003000-00015 pmid:http://www.ncbi.nlm.nih.gov/pubmed/10731008
      OpenUrlCrossRefPubMedWeb of Science
    62. ↵
      2. Jacobsson J ,
      3. Timpka T ,
      4. Kowalski J , et al
      . Injury patterns in Swedish elite athletics: annual incidence, injury types and risk factors. Br J Sports Med 2013;47:941–52.doi:10.1136/bjsports-2012-091651 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23543425
      OpenUrlAbstract/FREE Full Text
    63. ↵
      2. Lai A ,
      3. Lichtwark GA ,
      4. Schache AG , et al
      . In vivo behavior of the human soleus muscle with increasing walking and running speeds. J Appl Physiol 2015;118:1266–75.doi:10.1152/japplphysiol.00128.2015
      OpenUrlCrossRefPubMed
    64. ↵
      2. Fiorentino NM ,
      3. Rehorn MR ,
      4. Chumanov ES , et al
      . Computational models predict larger muscle tissue strains at faster sprinting speeds. Med Sci Sports Exerc 2014;46:776–86.doi:10.1249/MSS.0000000000000172 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24145724
      OpenUrlPubMed

    Footnotes

    • Twitter @drnoelpollock, @shanekellypt, @MickGiakoumis

    • Contributors NP and BM led the study design, data collection and write up. NP, SK, BM, JB, MG and GP contributed to data collection. JL reported the imaging and wrote the imaging sections of the manuscript. BS led the data analysis. All authors contributed to drafting.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.