Preventing hamstring injuries - Part 1.

Preventing hamstring injuries - Part 1: Is therereally an eccentric action of the hamstrings inhigh speed running and does it matter?Anthony Shield 1, Simon Murphy 2

1School of Exercise and Nutrition Sciences and Institute of Health and Biomedical Innovation, Queensland University of Technology, Brisbane, Australia, and 2Medical

Department, Leicester City F.C, Leicester, UK

Hamstrings | Injuries | Eccentric | High speed running

Headline

Hamstring injuries remain a major burden in sport (1) de-spite enormous research efforts and the development of

apparently effective prevention (2,3) and rehabilitation prac-tices.(4) Interventions that work in amateur and lower levelsof professional sport may not be fully implemented in the morecrowded training and playing schedules of elite sport (5) andperhaps they work only in less fit athletes or their benefits are‘drowned out’ by other training methods? We suggest that be-lief systems and philosophies shared widely by expert coachesalso reduce adoption of some evidence-based practices.

van Hooren and Bosch recently argued that isometric condi-tioning may be more effective at reducing hamstring injuriesand more readily programmed into busy competition sched-ules than eccentric exercise.(6, 7) Here, we critique these ar-guments and offer some insights into an eccentric strength andsprint-oriented injury prevention program that has been em-ployed to good effect in the English Premier League. We donot claim that anecdotal evidence is in any way definitive. Nordo we claim that isometric training doesn’t work or that ec-centric hamstring conditioning is the only effective means ofprevention. Instead we argue that recommendations to em-ploy an isometric approach instead of an eccentric one arepremature.

Do the hamstrings act isometrically or eccentrically in

late swing?The argument for isometric hamstring conditioning is based,to a significant extent, on the authors’ proposal that the con-tractile elements (fascicles) of these muscles act isometrically,rather than eccentrically in the late swing phase of sprintgait.(6) Current biomechanical models suggest that hamstringfascicles actively lengthen in late swing (8) but van Hooren andBosch argue that stretch in the series elastic element (aponeu-roses and tendons) of the hamstrings is underestimated inthese models and that active fascicle lengthening is thereforeover-estimated.(6) As modelling is based partly on assump-tions, it is prone to error and we remain open-minded regard-ing this aspect of the authors’ argument. However, the likelyexistence of some error does not prove that hamstring fasciclesbehave isometrically in sprinting.

Unfortunately, human hamstrings defy direct observation insprinting, partly because ultrasound probes don’t provide clearimages of their fascicles during violent movement of the lowerlimbs.(9) So instead, the authors describe studies of quadriceps(eg. 10) and plantar-flexor (eg. 11) behaviour during smalland large amplitude stretch-shorten cycles of jumping. Ultra-sound imaging of fascicle and aponeurosis behaviour in thesemuscles is effective but, as the authors suggest, inferences fromvastus lateralis and gastrocnemius research to hamstring be-haviour must be cautious ones. Differences in stiffness be-

tween the Achilles, patella and hamstring tendons along withdisparities in pennation angles, gearing ratios and fascicle tomuscle-tendon unit length ratios all increase the potential forthese muscles to exhibit diverse behaviours. Furthermore, themovement speeds in jumping are nowhere near those of sprint-ing.

In small amplitude stretch-shorten cycles, fascicles of themedial gastrocnemius are quasi-isometric when active, withthe lengthening and recoil of the muscle-tendon units al-most entirely limited to the aponeuroses and tendons (Figure1).(11) So jumps of this amplitude essentially involve bounc-ing off our in-built ‘springs’ and are energy efficient as aconsequence. Large amplitude stretch-shorten cycles involvegreater ranges of motion, as observed when performing jumpswith bigger counter-movements (eg a deeper knee bend) andlength changes occur in both fascicles and the series elasticelements of the vastus lateralis in these circumstances (Fig-ure 1).(10) However, vastus lateralis fascicles lengthen pro-gressively less as stretch-shorten cycles increase in intensity

Fig. 1. Representations of the contractile element (CE), series (SEE) and parallel

elastic elements (PEE) and pennation angle (Θ) during stretch shorten cycles (SSCs).

In small amplitude SSCs the CE remains at virtually constant length while active and

the lengthening (a → b) and subsequent recoil (b → a) occurs almost entirely within

the SEE. In large amplitude SSCs, the CE and SEE both lengthen (c → d) until the

CE becomes isometric and the remaining lengthening is confined to the SEE (d →e). Shortening initially involves recoil of the SEE (e → d) after which both the CE

and SEE shorten (d → c).

sportperfsci.com 1 SPSR - 2018 | Apr | 25 | v1

Preventing hamstring injuries - Part 1.

(jump height)(10) and the author’s point out that a continua-tion of this trend may eventually see the fascicles remain iso-metric.(6) They also argue that contractile elements lengthenin these large amplitude movements because the forces are ‘toohigh’ to resist and they thereby characterise this lengtheningas a deviation from optimal behaviour. However, an alter-native interpretation is that contractile elements simply haveto lengthen when large ranges of motion are utilised. Further-more, no evidence is presented that human hamstring fasciclesbehave isometrically in sprinting and these claims currentlyamount to well-reasoned speculation.(6)

Having argued that hamstring fascicles are typically isomet-ric in the late swing of sprinting, van Hooren and Bosch (6)then suggest that aberrant fascicle lengthening may occasion-ally occur and that this may cause hamstring strain injury.They write that poorly coordinated lumbopelvic motion mayoccasionally force the hamstrings to actively lengthen. Thisis an important acknowledgement because strain injury is un-likely to occur during isometric contractions.(12) However, wesuggest this is also a major flaw in the authors’ argument andthat a rationale for eccentric training fits perfectly into thisgap. We contend that the damage resistance rationale for ec-centric training (13) holds whether active lengthening occursin every stride or only on rare occasions during sprinting. Asa consequence, the typical contraction mode of hamstring fas-cicles in sprinting may matter less than the authors suggest.

Is ‘damage resistance’ a function of training intensity

or contraction mode?Eccentric training may exert at least some of its benefits viaan increase in in-series sarcomeres and the lengthening ofhamstring fascicles.(13, 14, 15) Accordingly, hamstrings withlonger fascicles appear less prone to injury than those withshorter fascicles,(16) possibly because sarcomeres experienceless strain and are less likely to be over-extended. While vanHooren and Bosch,(6) contend that the effect of contractionmode on fascicle length is not clear, their argument is mademostly on the basis of quadriceps studies, which are incon-sistent. However, as of November 2017, eccentric hamstringtraining has been shown to lengthen hamstring fascicles in 9 of10 published training studies while 2 of 2 concentric trainingstudies have been shown to shorten them.(17)

The authors also suggest that any high intensity hamstringtraining may reduce injury rates, regardless of contractionmode.(6) For all we know this may be true, but this spec-ulative argument seems to ignore the force-velocity relation-ship of skeletal muscle. Isometric contractions simply can’t beperformed at the forces observed in maximal eccentric-only oreccentrically-biased resistance training, so if intensity matters,eccentric actions are the means to maximise it. Nevertheless,there is evidence that contraction mode plays an importantrole in skeletal muscle damage resistance independent of ex-ercise intensity.(13) For example, the original rat studies thatcompared eccentric and concentric training of vastus inter-medius muscles employed decline and incline treadmill runningat trotting speeds for 15 to 35 minutes per session over 5 days.The low intensity eccentric actions in decline running resultedin the development of more in-series sarcomeres than did theconcentric actions performed during incline running.(13, 18)In one of these studies, the extent of force loss and the shiftin the force-length relationships caused by 20 maximal (elec-trically stimulated) eccentric actions was markedly smaller ineccentrically than concentrically trained muscles and this isinterpreted as enhanced damage resistance that may also helpprotect against strain injury.(13) In humans, there is also ev-

idence that mild eccentric training (downhill walking) resultsin protection from eccentrically-induced muscle damage,(19)while mild concentric training increases the susceptibility tomuscle damage.(20) Timmins and colleagues (15) have alsoshown that knee flexor training with maximal concentric ac-tions results in rapid shortening of hamstring fascicles whileeccentric training lengthens them, despite long muscle lengthsbeing reached in both. As far as we are aware, there is nopublished research on the effects of isometric training on ham-string fascicle length. However, Pollard and colleagues (21)have recently found that high intensity training with the razorcurl, a quasi-isometric hamstring exercise, does not increasebiceps femoris fascicle lengths while the high intensity Nordichamstring exercise does (unpublished observations). Thesecollective findings suggest that eccentric actions induce morefavourable architectural adaptations than concentric and iso-metric ones, regardless of contraction intensity, although morework on isometric training needs to be done to confirm this.

Impact on running coordination?Departing from their contraction mode argument, van Hoorenand Bosch propose that the Nordic hamstring exercise doesnot alter hip and trunk muscle coordination in running.(7)However, it has never been argued that this exercise altersrunning technique. It might also be argued that the authors’suggested alternatives, many of which are exercises performedfrom a Roman chair at neutral hip angles,(7) are also disparatefrom running in terms of movement pattern, velocity and pos-ture and that none of them seem particularly likely to directlyinfluence sprint technique. Nevertheless, the argument for em-ploying hip extensor exercises for hamstring injury preventionis one with which we entirely agree.(14,17)

Scheduling eccentric hamstring conditioning in elite

sportVan Hooren and Bosch contend that because it causes lesssoreness, isometric training will fit better than eccentric exer-cises into the crowded competition schedules in elite sport.(7)

Fig. 2. Variations of eccentric strength training days (red rectangles) based on

differing weekly structures. Balls represent match days (MD), MD+1 is 1 day after

match day; SSG) is a small sided game; BTB is a box to box game (the edges of the

penalty boxes become the goal line). In example 3, players who start the match do not

perform eccentric strength training. Substitutes who have played under 30 minutes

and unused substitutes perform a combination of eccentric strength training after on

pitch running.

sportperfsci.com 2 SPSR - 2018 | Apr | 25 | v1

Preventing hamstring injuries - Part 1.

Fig. 3. Example of lower limb eccentric strength training after a post-match

running session for substitutes. 3.1) 2 sets of 3 Nordics, 3.2) 2 sets of 8 double leg

Romanian deadlifts, 3.3) 3 sets of 6 K-Box squats 3.4) 2 sets of 8 double leg slideouts.

Fig. 4. High speed running drill to target peak speed at the end of training. Some

players will perform 2-4 repetitions depending on their volume of high speed running

in the training session.

This congestion is arguably highest in elite European soccerwhere teams may occasionally play as many as eight matchesin 21 days. However, the well-documented repeated bout ef-fect ensures that supramaximal eccentric training can be per-formed with limited soreness after the initial 2-3 weeks (2, 14)and one of us (SM) has experience implementing a hamstringinjury prevention program in the English Premier League with

a degree of success. We therefore believe that careful planningcan overcome these issues.

The incorporation of eccentric training into Leicester CityFC’s hamstring injury prevention philosophy was predomi-nantly driven by the growing evidence in favour of this ap-proach.(2, 3) Due to the revolving door nature of a multina-tional EPL squad, the club is acutely aware of the associatedcommunication barriers and differing training histories clubsmay experience with their athletes. As such, the strengthdevelopment culture has been based on avoiding technicallydemanding exercises (eg Olympic lifts) and thereby enablingplayers to push high loads for few repetitions. High intensitylow volume eccentric training of the hamstrings fits well intothis philosophy which is complemented by a broader, multi-modal approach of weekly load management and high speedrunning exposure.

While muscle soreness often limits the application of eccen-tric training, perseverance allows soreness to diminish overtime. Players are encouraged to view soreness as an adapta-tion rather than a concern, and to trust staff to protect themvia appropriate manipulation of loading strategies. Timing ofeccentric work in the training week therefore becomes crucial.Eccentrics can be safely administered following adequate re-covery post-match, but must also permit a suitable period forrecovery ahead of the next game. At Leicester City eccentrictraining has traditionally been placed three days post-matchday (MD+3) for a one game week, but this approach must beflexible to accommodate change (Figure 2).

Where periods of high fixture congestion occur, the focusturns to maximising the windows of opportunity for playerswho don’t play or play reduced minutes from the substitutes’bench. For starting players, the focus in this period is solelyto recover from the accumulated match day muscle damage intime for the next fixture and eccentric training is not appro-priate. Portable training equipment can be used both homeand away to achieve this stimulus in the target group after thematch. This is complemented by short sessions with a highspeed running focus on the pitch.

Leicester City employ knee and hip orientated hamstringexercises. Whilst there is an appreciation of individual mus-cle activation for the chosen exercises,(17) this doesn’t exclu-sively guide the process as some exercises are favoured fortheir simplicity. There is a focus on building an appropriatebase level of strength via an overload programme in presea-son. Doing so builds eccentric strength levels and presumablyevokes favourable architectural adaptations(14, 15) that canbe maintained with a relatively low volume of high intensitywork throughout the season. In season training typically in-volves 1-2 sets of 3 repetitions of Nordics, 2 sets of 8 doubleleg/weighted eccentric slideouts and 2 sets of 6-8 double orsingle leg dumbell Romanian deadlifts (Figure 3). The clubcan objectively track knee-flexor strength but recognises thatthese low volumes may limit strength improvements. Insteadthe maximal Nordics are less of a strength training set butmore a monitoring/structural training set to offer an eccentricexposure that may maintain favourable fascicle lengths andprovide strains to the series elastic element. The added hipand knee volume sets then top up this exposure and help con-dition the muscle in a less strenuous manner. All exercisesare performed after training to work the muscle in a fatiguedstate. This approach aims to make weak players strong, butmaking strong players stronger with large external loads isn’tnecessarily the priority. The club feels that this approach helpsavoid potentially inhibitory levels of soreness that negativelyinfluence compliance.

It should be acknowledged that strong athletes still injuretheir hamstrings. We suggest high speed running is the most

sportperfsci.com 3 SPSR - 2018 | Apr | 25 | v1

Preventing hamstring injuries - Part 1.

Table 1. Summary of the major arguments addressed in this paper. These are not exhaustive and the reader is directedtowards the van Hooren and Bosch papers for a more detailed argument.

The argument for isometrics The counter-argument

Hamstring fascicles are argued to be isometric in the late-swing of sprinting & isometric strength training is thereforemore specific.

All currently published modelling studies show activelengthening in late-swing but these models do involve as-sumptions. van Hooren and Bosch acknowledge that eccen-tric actions may occasionally occur inadvertently and causeinjury. We suggest that the argument for eccentric trainingis valid regardless of whether the hamstrings act eccentri-cally in every stride or only rarely.

The benefit of hamstring injury prevention exercise is deter-mined by contraction intensity more than contraction mode.

Only eccentric actions have significant potential to causemuscle damage at moderate muscle lengths and triggeradaptations that create damage resistance (eg the repeatedbout effect).

The NHE does not improve coordination of hip extensorand trunk muscles the way that other exercises do.

Improved coordination is not the mechanism by which theNHE works. Whether isometric exercises are more specificto running remains to be seen. We agree that hip extensorexercises are important in injury prevention.

Easier to schedule isometric than eccentric training intocrowded training programs because the former causes lessdelayed soreness.

This is undoubtedly true but only of relevance if isometricsprove effective. It is also possible to program low volumes ofeccentric strength training into a crowded training schedule.

valuable injury prevention tool we have while it is also thebiggest contributor to hamstring injury causation. Throughcareful planning the club aims to mitigate risk by reducingtraining pitch dimensions for 72 hours after games whilst mus-cles are still recovering from the match day insult. In doing so,players are restricted by space and opportunity to reach highspeeds and are therefore protected from the common mech-anism of hamstring injury. As the week progresses and theplayers are better recovered, training pitches increase in sizeand allow players to reach higher speeds, which are thought tobe an important exposure in the training week. We acknowl-edge that football is variable and the flow of a training gamemay not require some players to achieve this exposure. As aresult, the club includes a high speed running drill (Figure 4)at the end of the session to train this variable in a controlled,fatigued state. We argue that sprint training in this controlledenvironment can have protective effects regardless of terminalswing contraction mode.

Whilst isometrics are a key part of the club’s rehabilitationphilosophy, they don’t yet feature as part of the hamstringinjury prevention programme. While acknowledging the po-tentially positive changes that isometric training may evoke intendon stiffness and collagen at the musculo-tendinous junc-tion, the club prefers the limited time spent with the playersto be evidence led.

Comparing the current evidence for isometric and ec-

centric hamstring conditioningvan Hooren and Bosch correctly point out that there is noevidence that eccentric conditioning is better than isometricapproaches. As far as we are aware, there are no experimentalstudies examining the effects of isometric hamstring trainingon injury rates and a large scale (n = 500+) randomised con-trolled trial comparing these methods would be an ideal testof this argument. We acknowledge that even if isometric ap-

proaches prove to be half as effective as eccentric approacheshave in the past,(2, 3) they would likely have a greater impactin elite sport as long as their adoption is adequate.(5) Fornow, however, there is level 1 evidence, including systematicreviews (22) and large scale randomised controlled trials foreccentrics,(2, 3) while the evidence for isometrics remains atlevel 5 (expert opinion). Indeed, the argument for isometricsis, for now, entirely theoretical. So where does this leave thepractitioner who designs training programs? Not all decisionsin sport can be evidence-based and not all findings from re-search fit neatly into the schedules or the beliefs of any giventeam. We offer insights into how eccentric hamstring condi-tioning can be accommodated in elite sport despite crowdedschedules; however, we also acknowledge that expert opinionhas a role in the development of injury prevention strategies.Future research may well provide evidence for isometric ap-proaches to hamstring injury prevention and we suggest thata skilfully programed combination of knee and hip oriented(14, 17) eccentric, isometric and conventional resistance train-ing approaches may be used, along with high speed running,with likely success in elite sport. Finally, we commend vanHooren and Bosch for publishing a thorough and well-writtenargument for isometric hamstring conditioning.(6,7) We be-lieve, like Popper, that “the growth of knowledge depends en-tirely on disagreement” and we hope that our arguments areaccepted in this spirit. We also look forward to future scien-tific exploration of isometric approaches to hamstring injuryprevention and enhancing athletic performance.

AcknowledgementsTony Shield is a shareholder in a company that makes a ham-string strength testing device which measures eccentric andisometric knee flexor strength. The device is impartial to theresults of the current debate. Simon Murphy has no conflictsof interest to declare.

sportperfsci.com 4 SPSR - 2018 | Apr | 25 | v1

Preventing hamstring injuries - Part 1.

Twitter: Follow Tony Shield @das shield and Simon Murphy@SimonMurphy23.

References1. Ekstrand J, Walden M, Hagglund M. Hamstring injurieshave increased by 4% annually in men’s professional football,since 2001: a 13-year longitudinal analysis of the UEFA EliteClub injury study. Br J Sports Med. 2016;50(12):731-7.2. Petersen J, Thorborg K, Nielsen MB, et al. Preventive ef-fect of eccentric training on acute hamstring injuries in men’ssoccer: a cluster-randomized controlled trial. Am J SportsMed. 2011;39(11):2296-303.3. van der Horst N, Smits DW, Petersen J, et al. The pre-ventive effect of the Nordic hamstring exercise on hamstringinjuries in amateur soccer players: a randomized controlledtrial. Am J Sports Med. 2015;43(6):1316-23.4. Mendiguchia J, Martinez-Ruiz E, Edouard P, MorinJB, Martinez-Martinez F, Idoate F, et al. A Multifactorial,Criteria-based Progressive Algorithm for Hamstring InjuryTreatment. Med Sci Sports Exerc. 2017; 49(7):1482-92.5. Bahr R, Thorborg K, Ekstrand J. Evidence-based ham-string injury prevention is not adopted by the majority ofChampions League or Norwegian Premier League footballteams: the Nordic Hamstring survey. Br J Sports Med.2015;49(22):1466-71.6. Van Hooren B, Bosch F. Is there really an eccentric actionof the hamstrings during the swing phase of high-speed run-ning? part I: A critical review of the literature. J Sports Sci.2017a;35(23):2313-2321.7. Van Hooren B, Bosch F. Is there really an eccentric actionof the hamstrings during the swing phase of high-speed run-ning? Part II: Implications for exercise. J Sports Sci. 2017b;35(23):2322-2333.8. Chumanov ES, Heiderscheit BC, Thelen DG. The effectof speed and influence of individual muscles on hamstringmechanics during the swing phase of sprinting. J Biomech.2007;40(16):3555-62.9. Cronin, N. Free communication Hamstring strains in highspeed movements: Implications for injury prevention. 21stAnnual congress of the European College of Sport Science.2016. Vienna.10. Ishikawa M, Finni T, Komi PV. Behaviour of vastus later-alis muscle-tendon during high intensity SSC exercises in vivo.Acta Physiol Scand. 2003;178:205-213.11. Kawakami Y, Muraoka T, Ito S, Fukunaga. In vivo mus-cle fibre behavior during counter-movement exercise in hu-mans reveals a significant role for tendon elasticity. J Physiol.2002;540(2):635-646.12. Lieber RL, Friden J. Mechanisms of muscle injury gleanedfrom animal models. Am J Phys Med Rehabil. 2002;81(11Suppl):S70-79.

13. Lynn R, Talbot J, Morgan D. Differences in rat skele-tal muscles after incline and decline running. J Appl Physiol1998;85(1):98-104.14. Bourne MN, Duhig S, Timmins RG, Williams MD, OparDA, Najjar A, Kerr GK, Shield AJ. Impact of the Nordichamstring and hip extension exercises on hamstring architec-ture and morphology: implications for injury prevention. BrJ Sports Med. 2017a;51(5):469-477.15. Timmins RG, Ruddy JD, Presland J, Maniar N, ShieldAJ, Williams MD, Opar DA. Architectural changes of the bi-ceps femoris after concentric or eccentric training. Med SciSports Exerc. 2016b; 48(3):499-508.16. Timmins R, Bourne M, Shield A, Williams M, LorenzonC, Opar D. Short biceps femoris fascicles and eccentric kneeflexor weakness increase the risk of hamstring injury in elitefootball (soccer): a prospective cohort study. Br J SportsMed. 2016a;50(24):1524-1535.17. Bourne MN, Timmins RG, Opar DA, Pizzari T, Ruddy J,Sims C, Shield AJ An evidence-based framework for strength-ening exercises to prevent hamstring injury. Sports Med.2017b doi 10.1007/s40279-017-0796-x18. Lynn R, Morgan D. Decline running produces more sar-comeres in rat vastus intermedius muscle fibers than does in-cline running. J Appl Physiol. 1994;77(3):1439-44.19. Nottle C, Nosaka K. Repeated bout effect conferred bydownhill backward walking. J Exerc Physiol Online. 2005;8(1):1-10.20. Gleeson N, Eston R, Marginson V, McHugh M. Effects ofprior concentric training on eccentric exercise induced muscledamage. Br J Sports Med. 2003;37:119-125.21. Pollard C, Bourne MN, Williams MD, Timmins R, OparDA. The impact of exercise selection and training intensityon biceps femoris long head architectural adaptation. Unpub-lished Honours thesis, Australian Catholic University, Mel-bourne. https://drive.google.com/open?id=1pazSfGizEFAcafNfCOOZFuLjNPIJiq222. Goode AP, Reiman MP, Harris L, DeLisa L, KauffmanA, Beltramo D, et al. Eccentric training for prevention ofhamstring injuries may depend on intervention compliance:a systematic review and meta-analysis. Br J Sports Med.2015;49(6):349-56.

Copyright: The articles published on Science Performance and Science

Reports are distributed under the terms of the Creative Commons Attribu-

tion 4.0 International License (http://creativecommons.org/licenses/by/4.0/),

which permits unrestricted use, distribution, and reproduction in any medium,

provided you give appropriate credit to the original author(s) and the

source, provide a link to the Creative Commons license, and indicate if

changes were made. The Creative Commons Public Domain Dedication

waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the

data made available in this article, unless otherwise stated.

sportperfsci.com 5 SPSR - 2018 | Apr | 25 | v1