A biomechanical analysis of racewalking gaitMARILYN A. CAIRNS, RAY G. BURDETT,JEFFREY C. PISCIOTTA, and SIffLDON R. SIMON

GaiI .Analysis Laboratory, Childrm's H ospitcl,Boston,MAAZlIS

cAIRl,Is, M. A., R. ".

t.ffi5c n$clorrA" and s. R-SIMON. A biomechanieal andysis of racrrvalking gjeit Med. Sci-Sprts Exqc., Vol. 18, No.4, p'p.446-453, 1986. The purposc oftisstudy !a$ to investigate ths kinetic and kinernatic propcrtie of thelorrer extremity durirry the phases of the rwualkirg gait as definedby the rules and compare ttr€se to normal walki$ and ruaningCine.matogrgphic and force plate data werc gatherEd from l0 com-potitivc racoualkers underfour gail conditio$. Temporat Fram€*ert"angular displacements of the lower extremity, ground reaction forces,patb of thc center of pr€lisur€, and the joint forcs during the fourgait conditions were corryrared $ing a repeatd measures analysis ofvariane.

Racwalkcrs exhi6;1sd significanrly inrreased (P < 0.0t) mafimalankle dorsiflexion, maximal knec extension, angular di5gacernentsof the pelvis, rnedial goundtoot reaction'forces, peak plantar flexionmomeat, and ertemal peak knee hyperextcnsion momert rhaa didrralking or running Based upon the results of this inr,rsiigation, itwas deteflninsd that rhe git of racuratking exhibits some biom*chanical charactrri$tks,which arcdiffercnt ftom ttre gair of wdkirrSor running

BTOMECTIANICS, GAIT ANALYSI$, RACEWALKING,KINETICS. KINEMATICS

It is clear ftom a literature review in the area ofhuman locomotion, that the kinematics and kinetics ofnorrnal walking and runningat a variety of speeds havebeqn well documcnted U,2,5,I0, 13, l7). Despite itsEuropea.n popul4nty and the prcs€nce of racewalkingin the Olympic Games, there is little published researehdocumenting the mechanical or physrolqgical charac-teristics of raqewalking Franklin et al, (8) repor_ted thathighly trained racewalkers and marathon runners havesimilarphysiological and psychological profiles. Menierand Pugh (12) and Marchetti et al. (l l) repo:tcd race-walking lo be a less efficient ,forrn of locomotion thanrunniug at velerities greater thaa I knq'hr-r. Cavagnaand Franzetti (3) reported racewalking at hig[ speeds10 be less expensive in net energy expenditure thannormal.walking- Payne (15) reported the effecfs of therestraints ofracewalking gart on the forces evoked fromthegfound by the athlete as compared to normal walk-ing while Fenton (7) compared the ground reactionforces of elite and compelitive racewalkers. Murnay etal. (14) repo-rt€d temporal components as wEll as the

Sshmitcd for.nftlicatbn MaJ, 1985.* oeptc,ll br n,btietln MadL rgE6.

0rs6-9r3t/86fl 804O446S2.m/0r,G[iorrE AND SCETSE r!,r SPOAT$ ANO SGFCISECsryiqftto lget by tteAmetican eoFqFof SporigM€.fci*

vd. 16. ib-.3Firn|gdrr Ug.^

displacEmmt Ftterns of body segments of olympicraccwalkers.,Ptrillips ad Jenson (16) reportd similarjoint displacement pattelns with elite racewalkers whil'eDouglass and Garrctt (6) found stride length to be amajor contributor to velocity among junior elite rac+walkers.

The gnrt employed in racerualking is a ftrnction ofthe rules which goye.rn the sport. In order to exegute alegal gait, constant contact with the ground must bemaintaioed at a[ tines, and a double support periodmust be established during each g5rcle. The rules asstated in the International Amatatr Athlaic FederationHandbook (9) also requi:res the supporting lqg to be*straigfitin the vertically uprieht position." Tho restric-tions of the nrles of racerlralking result in a gart patt€rnwith some unique biomechaniel characteristiqs" Theptesent study uas desig[sd to dqcUment the kineticand Hgenatic propcrties of the lower extrenity duringthe phases of the racerralking gait as defined by thsrules and cornparc thw to normal walking and run-ning

i

iMEIHODS l

Subi€cts. Ten oomperitive ramwalkem, two femalesand eiglt males, volunteered to seryc as eub,jests forthis study. The .subjecti' self,reported performancetimes over l0 knr averaged 5O:18 with a r:rngp of 46:25ts 55:14. Six of the subjqs wer€ prrrently or had inthe past bqeq nationally competitive.,hior to partici-pation in the study, written informed consent wasobtained from each zubject.

Protocol. Each subjcci rworted to the Gait Analy$skboratory Childrgos' Hospital, Boston, for a datacollection sessioa. The subjeOts ncre preparod accond-ing to the protocol described by Simon et al. ( t 8). Pteoesof black rape, 2.5 wrz,with centrally locatd white dotsrcreplaced onthe following po-ntionson both sidm ofthe body: the dorsum of the foot between the first andsecortd nelata$als; the laterat aspect of the foot atapproximately the calcanoocuboid joint; the lateral as.pect ofthe fifth meraa$a! phalalge! joinq the lateralmalleolus; the anterior approximation of the tibial telar

46 I

itI.,,1

A BIOMECTIANICAL ANALYSTS OT RACEWALKING GAIT

joint; lhe patclla; the lateral aspErt of the knee joinUthe anterior superior iliac spjne; the latenl aspect of theneck at aF$pximately the level of the sixth cervicatvertebra; the tip of the acrorflion prtloess; the lateralep.icondyle of tlie elbow; and the dorsum of the wrist.A,niarker was also placed on the super-ior margin of thesternum, and 25 cm rods with red tips werE fixed byrreans of velcm anachments and anchoring tape stripsto thesacrum, anteriorfemur, and anterioitibia Thesesticks werc used to assess rotary motion of the pelvis,:fernur, ald tibia-

Motion of the trunk and limb segments wils recodedat 50 frames/s by three Photosortic hieh*peed 16 mmsameras niounted on steel posts secured at a height of:32 inches from the lloor and phgse-locked for synchro-nization. The cameras w€re positioned similar to thosercportd by Suthcrland and Hagy (19), 12 feet ldteraland 39 feet anterior to the force plate. A strain gaugeforce plate having resonant frequencieS maintaind at500 Hz was usd to record the ground reaction forcesundet the right foot. This force plate was solidlyrnounted into the floor of a 4O focit runwal with theplate's surfacp at the level of the surrounding floor.FooJ-hr reaction forces in the vertical" anterior-pos-terior, and medial-latcral directions wer€ recorded andstored in the computer data file, To determine the pathof the center of pressule, paper was placed over theforce plate andan impression of the subject's fost wasobtained during the stancr Fortron of the gait cycleutilizing charcoal dust on the sole of the sloe. .

After practice trials, each subject was filmed" andforce plate data werc collected for at least two trials ofeach of the fotlowing gaits: a.normal self-selected velcc-ity walk; a racewalk at a self-selected training Fcq aracewalk at a self-selected corhpetitive pace; and ajogat a rdocity approximately that of the competitiveracewalk. After being developed, the film was projectedonto a V4qguard Motion Aaalyzer. For each of thethree camera views, the twodimensional coordinatesof the designated anatomical landmirks and the footstrike events of one cornplete gait cycle for each trialwere. entered into the computer using a Graf Pen SonicDigitizer operated by a trained technician. After thetwodimensional projected coordinates of each pointwere recorded, a computer program translated the co-ordinates into the threedimensional coordinates usedto define segxnent locations during the gnrt cycle (18).The center of prcssure coordinates werc calculated foroach rrial with rcspect to a shoe coordinate sy$temoriginating at the heel with one axis along the shoemidline.

Data calculatons. Frorn the cornputer-stored filmdata, the following parameters of each gait.cycle pcrecalculated: velocitg cadence; stride length; joint angles;swing time; stanc€ time; and joint velocities Internaljoint moments {F!g l) were calculated in the saeittd

Iigrire l*Interorl irlinl momcnts et lhe r*le (+M^ = pluilarllexbo), Lnee (+Mr = efteffim),ud hip (*Mg = extelskr) in tbeisgitEl pline ad the hip (+Mal = rtductbn) irt thc fmdrl Dlrre.Foraes iuclude Ra Rvr Rx = vcrticrl, anterior-pqst€rlor" ud nedhl-hterel compoireirts,of lhe'ground reaction lbrce Kz, Kv atrd Ha Hv= reqction,forctx ef the Lnee and hiD; rff* ft = reigbt ofthe ihrnlrqil thig!. Cqordluates hclnde ZD Y3 Xl = 4l,X coordnetes ofthc cetrtcr of pressure Of the forces of ttc fooq Za,Y7 = 7aYooordiqtcs of th htcral udleolnE Zs, Ys, f,5 = t Y, X e*rdinefetof tlie ceater of mes of the slul.; Zr, Yt = 4 Y coordtnrter of thebteral trcc nnrte6 Z4,Ya,Xa = ZrY, X coodiirrtes of tle c€oaerof nrss of the thie|l4Ze* Y". &t = Z Y, X osordioates oJ the gprterfochalter.

plane about tAe ankle, knee, and hip joints and in thefrontal plane about the hip joint. The mass of the footwas neglected in the ankle momcnr calculationq butthe inertial properties and accelerations of the shank,foot.and thigh wete included in the sagittal plane kneeand hip mom€dt caliulations. The followilg asump^tions were made in the calculation of momenB in thesasttal plane:

Av=RvAz: Rv

Kz: Az - Ws - ms.%

Kv: Av - ms'Ys

Hz=Kz-Wr -mt'ZrIfY: KY - m'Yr'

447

/t(o,=n,>S,h.4c

IR7

{r{,4r,J.-"\TY."

",-+ )) ,Y.4c

' 1'*Kz

Az

MAg

Ct

,fJ

448

The aoceleration of the leg was ngglectd in the frontalplane in calculating the abduction-adduCion momentat the hip.

The following cquations were used to calculate theinternal joint momenn at the ankle (M"), knee (MK),and hip (Mn) in the sagittal plane arid the hip (M*s) inthe frontat plane:

Me = Rz(Yn - Ye) * Rv(Ze * Znl

Mx : Az{Ys - YJ - Av(Zs - Zn)

- Me * Isds

Mu = Kz(Yx - Yr) + KY(ZI - Z:r)

+ Hz(Yr - Yd * Hy(ZH - 7'r,

- Mr * Irqr

Mnn - Rz(Xu - Xil - &x(Zu - Znl

- WI(XH - Xr) - Wr'(XH - Xr)

- Ws(Xn - Xs)

where Rzo Ry, Rx ar€ the vertical, an&rior'posterior,and medial-lateral components of the ground readionforpe; Ws, Wr aie the weights of the shank and thieh;and Iscs; Ilcrar€the momeirts gfinertla of thesrymenl(shank and thigh) times the ameieration of the s€gment.

Data analysis. The data analysis was orgarrhed intothrw areas: (l) temporal evaluation of selectd eventsthrough the gait cycte; (2) angular analysis of the lowerexfemity at the occrrrrence of scldcted events of thegart cycle; and (3) gound reaction foroes and jointmoments of the lower extremity calcutated at fhe oc-currence of selected er€nts. Means and standdrd devia-tions were calculated for each data sel, and the $gnifi-cancd of the difference be,tween the means of eachvariable across conditions was tested using a repeatedm€asures one-way analysis of variance. Tukey's hon-estly signilicant difference suilistiC was usd to bolatethe sourceof the difference in each case.

RESUTTS

Speed, €dence, stride length, stane time, and swingtime were determined for t0 subjects during the fourgait modes. A summary of the mean temporal compo-nents is presented in Table L The mean Velocities oJracewalking and nrnning were not significantly differ-ent from each other and were about nryice that ofwalking

Angular displacements were determined from thefilm for the lower extrcmity sEgpents of the rieht tegduring the four gait canditions. Mean m4ximal angulardisBlacements and exctrrsions weie determined and

MEDICINE AND SCIENCE IN SPC'RTs AND E"XERCISE

TABI"E 1. Tenpod curponene d, galL

ConpotFrtCotrpcfitue'Recra&

Sptid (m. sec-t)Cdene{sffies-

qoco)Stldelengh(m)Sttreltne (sec)

Swir€ time{sec)Sbncelimelsairlg

lirn€ralio

r.$*024r.0{ r 0.09

116+ 0.17061*007038 i0.031.50 i 0,15

2.89 +039t.30*0.(p

2.'r3 * 0.200.42 * 0.06o36*0.011-15*.Q,n

3.63i0.151.54 * 0;14

237ton0.31*0"06031t0.02't.e-''0.19

3.02 +O53r38 *0.14

2.75*4480.31*9.050.45 * 0.04

0.70 + o12

TABTE2. l@r ftodrd deviationsof a*le, ldtee, hit,srd pelyis.

Cmd[to0

JoFlAqtloCqrpeffiFsm[

Ankle dmfilexiot{110:'

Ar*le dadarlfexkn(110"1

lc|eotlexir|{1801tgpextersirl(f)lhflexil'gll{p€dsrsi(nfl8f}PdlticdtPdvico0rq'JtyPehi;tEtalion

11.8*.52 U..49t23 26.90+73' ?f,34*9.1

28.08+5.0 n.ffit77 2473*,9.9 24.e,*.63

6928 + 3.9 ?3.m + 5.7 71.51 t 7 .1 91 35 ! r t.,r'181.9 + 23 1g/.06 r 4.9f rC/.90 * 6.8t 10851 + S.8453114.9 55.34*5.8 €014s75' 5268*6.r$a t7.9 13.42*45 10,30*2.4 925i2,74.99+3.tr 658*33 7.Zl*.38 7nt1.l

1824*31 23"5/*3.8t 2425i4.61 14.93*38nn *8.1 33.98 r 4.Gt 35.70 + 8.4t 16.m i 5-0

' S@rtf tfftercrn hsn otur wfrps; P < 0.01.

t SiilFatV Clfferent ftun sE* a$ tut rdues; P < 0.O1.

compared across conditions usiag one-way analysis ofvariance with repqated measur€s (Table 2). During thecomp€titive speed of raewalking the ankle achievedsignificantly geater maximal dorsiflexion just prior toheel strike than during the other gaits. At the knee,there was significanily greater maximal'extensibn to aposition ofhyperextension during the stanoc phase ofraccualking as comlxrr€d to walking or running Theraeewalkersin this study displayed an increasodamountof hip flexion during the swing phase of raaeualking,which was significantly greater at the competitive ve-'locity as mmpared to the olher gart conditiong Themean excursion of pelvic tilfing in the sagrttal plancwas signifisantly greater during racwzllkingthan duringwilking Pelvjc obliquityin the frontal plane and pelvicrotation in the transvcrse plane were significantlygrcater during the racewalking gaits than during walkingor runnirg

Mean peak grouird reaction foroes between the shoeand floor expressed as multiples of body weight weredeterrnined and compared acnoss gait conditiom usingone-way analysis of variance with repeated measures(Table 3). The vertical ground-foot reaction force in-cr€ad sienifica8tly across gait conditions from walkto racewalk to ruh. The anterior ground-foot reactionfoie displayed during run:dng ruas significanrly grealerrhan during the other gait conditions. The medialground-foot rcaction force3 during the racewalking

A BIOMECHAN|CAL ANALYSTS OF RACEWALKINC GAIT

TABTE 3 lrean ped( $ond rHrtim tuces{x boo r'yeirl"/t} in $tce.

44.9

TABI.E rf. lrean Fint nfiErrb as p0roQnt d Fody ue'gfrt x loiller o(bgrtlylengti.

qrilrdn$oofille

CompettYeR*era& But TfihlS

nrcilrNlrColFettwRrccr* Rrnvsricd t.31 *o.tr 1.$,!02tr

Anbrid 023{-0.04 026-10.04Fosl€rhr 0.31 + 0.06 0.36tl).r0lredd 0.10+olx 0.11+0.04tl-a_r€t"d 0.(tr*0.@ 0.07*0.05

1.65 + 0.i15. e{5 + 0.1'{'0.29 * 0.07 0J4 + 0.08'0.35s0,13 0.34 * 0.090.1610.06t 0o8t0.{x0.0tr0.05 {xrs0.04

' $grlficanty dletent fidn all ohs rr*lq P < O0r.f Si$ficadydtrerentfrorilrdk adnn vdues;P< O01.

ga+ts rlere significantly larger than d"ring walking orrunntng.

Mean internal joint moments werp Galculated in thesagiJtal plane about the ankle, knee, aud hip joints andin the frontal plane alout the hip joint. Peak jointmom.ents and joint mor4ent$ at specific events wer€norrnalized by body weight and lower extrcmity lengthand compared acrossconditionr using one-way analysiiof variane with repeated measqres (Iable 4) At theankle, the peak dorsiflexion moment was significantlygreaterduring racewalking than durirry walking or run-ning. The peak plantar flexion moment increasd sig-nificantlyacross garts from uralk ts racewalk to run. Atrhe knec joint, the peak knee :flexion moment wassignificantly gDeater in raoewalking than in running orwalking; The peak kne extension moment was signif-icantly g@tat in running whereas the peak knee hy.percxtcnsion moment was significanlly gea-ter in,race-walking as compar€d to the other gait conditions. Theh1'pereirtension moment rnas defined as oaurring whenthe extenal moment about the knee was in ttre dir€c-tion of extension and the knee joint was extended tomore than 175". The peak hip abduction momeat ofracewalking and nmning was significantly greater fhanthat dispJayed during walking. The peak hip aMuctionmoment of the competitive vclocity of racewalking wassignificantlv geater than that of running

DtscusstoN

The results of rhis study indicate that there are someunique kinetic and kinematic properties of the lowerextremity during the nacewatking gaitas defined by therules. The selFselected velocities of the two speeds ofracervalking were 2.89 and 3.63 m,$-r. Murray et al.(14) rcported self:setected velocities of 3.25 and 3.44rrl-s€c-r for ttvo olympic raccwalkers. As speed is in-creased from normal rvalkirrg to racewalking there ar-e

some temporal and angular characteristics which arebiomechauically related to increases in velocity whilemaintaining double,supporL Avcrage durations of thest4nce and srvipg phase for the l0 subJects expressqd intime and percent of the gart clele iare prcscnted inHgure 2. tn racer'r*atking the percent ofthe cycle spentin the stanc€ phase decreased, and the percenl4ge of

&,09f9.66t

1?.17t13.t58J1t

1.9

21.9421241n

30372659

1r.13t 0.q1

1.78 1.17

u,76 14,8325.16 20111.76 't.g'34.07 n.W30.s. ?{.81

'$bnifcildydifHeot l?qn al atn,r vahres P<0.01.t Siltifcilrty dfiE eil fruti nal( ad nn vdfs; P < 0.01.

the cycle spent in tbe Cwing phase incteased. so that, atthe comlEtitive velocity. the stane and swing phaseswere nearly equal. The double limt support phaseshortenedto approximately 0.01 t, wiJh several subjectsshowing periqds of ftoatftrg as a result of toe-offbeforcthe opposite heel strike. These periods of floating werclimited to lessthan two fhasres orO.04 s and would nfibe vizually discernible. Velocity increases in racewalk-ing ascompargd torunniqg are largely dw€ndent uponincreasingcadelceonce ma:rimal stride length has beenaehierred due to the necessity fix double support.

At first glane, the racewalkers' motior loo*s awk-wand, exaggeralpd and unnatural. The rationale for thisseerningly exa3gerated motion has an anatomical andmechanical basis, as raCewalkem maximize the urc ofthe lever system while conserving mechanical enegy.At the anklen increaspd dorsiflexion appeared just priorto heel strike as the subject reached foruard for maxi-mal stride length. Incr,eased plantar flexion of the hcklimb toeEther'with increased knee extension and dor-siflexioa oJ the forward limb were also identified byMurray €t al. ( 14) as a rneans of frrnctional lengtheningresulting in forward projection of the calcaneus andincrcased stride tength.

ln normal walking the alternating uave of kneeflexion and extension rg?resents the primary mecha-nism which controls the vertical excursions ofthe c€nterof gravity. During racewalking this wave ua eitherabeeni or diminished The racewalking rules rpquire thestraight knee position during the stance phase whichhas the effect of abruptly raisitrg the mass cen. ter at 4time whe.l it is already risine. Dwing racewalldngsubjects spent 44 to 57Vo of sttance with the knee inhyperextension (defind in this aoalpis as grearer than1751. Motions of the pelvis:in the frontal plane dwingracewaltingtend to minimize the vertical excursipm ofthe mass enter. The h4r incrcase ia pelvic obliquityduring racewirlking is a result of a larcral drop of the

Peakdattrtexigr 16.9fPed(dAr!fllexbfi 7.nftaktne flexin 9.58PBakfffieqtc|slfi 15.92

P€li|trtFhlpelergt! c22sm

l{rpelleriontliilrst+ 1,33

Pq€iterriq|PeakFnerirn{stsnoe} 19.78Pesf tb?xFFho 1686FBf Eadd'.dih lnPealrthabddfi* ?l.1trAvereebbl[unEn 20.fil'

a$t 8o.g93rt 4glfial 7.16r2fg 9.72'

MEDTCINE AND SCTENCE IN SPORTS AND EXERCISE

srANcE 6295 I SWING 3A% WALK t.a3 m.sCl

Figure 2-Arerege dunfioq of stlme andswing of l0 competitiye rooerdters duirgfom gzit cuuditions srDnrssd rs r peraert ofto{al cycle tirc of me lcg.

STANCE srt% | StflNG 46%

STANCE 50% | gwrr,le eo*

STANGE 41% | SWING 5S96

tFAI{['lc RACEWAIX 23g rn rac-1

Cgtv--lP€TTnvF BACEWAL( eOg rnlsdlt

RUN 3.61 m"sec '

liltl

pelvis away from the stanc€ leg. This latenal drop tendsto deerca$ the uprmrd excursion of the center of massof the upper body during the stance phase te compen-sate for the straight leg- Munay et al, {14) and Phiilipsand Jensen (t6) identified similar crmpeEsirtory re|:rc-tions to tlris tateqal pel"i" motion in the shoulder girdleand through th€ v€rtebral columr g motion of thepetvis in the transverse plane contributes to functionallengthedng of stride length as th€ pelvis on the swingside is rotated forrvad.

The raewdkers in fhis Study, used increasedamounts of hip flexion during the middle of the swingphasg. After achieving hip Ilexion, the subjects reversedinto hip ertension sooner during the racewalking gart$than during normal walking which displayeda gradualrevenal. Murray et aI. (14), in a study :corqiparingraceralking to fast normal walking, reported that nor-mal subjecs progress from frc to fast w4lking bylengthening stcps with increased hip flexion of thefonrard limb at heel strike. Raerralkerg however, uti-lized hip extension during late suring to galn mornen-tum to pull the body forward rather than restrainfornard motion at heel strike.

The avcrages of the ground reaction force curves brlO subjects arc shown in Figure 3. Since there was littledifference berween tlre ground reaction forces of thetwo racewalking gattq only the cnmpetitive velocityras€nalk is shorrn To ac*rieve loager strides and tosustain the non-support period of running an upwaldaccelenation of the c€nteJ of gravity requires a largervertical and anterior fope oomponent during stance.Although velocif increased aeross garts, posteiisrground r€action forces were simifar during the brakingphase. This is in contrast to:increased braking forces in

-4 .5 .6

3econds

.9 r.9

mceualki4g reportd by Payne (15). The magnitude ofthe anterior-posterior ground reactioil forces in thissnrdy are similar to those reported by Fenton (7),Fenton (7) also reported that tbe transition from porteriorly to anteriorly direstd force occurred early inracovalking; at 30to 40Vo of stance,as opposed tg 50%of sance in normal walking lhe avegge transition inthis study was at 4]7o of racewalking stance, The me-dial"lateral ground reaction force curves ofrunning andracewalking are similar in pattern early and late in thestance phace. Payne (15) rccognized a similar genenaltrend in the medial-later4 force amplitudes of rair-walking and mnning cxaept for the marked incrsaseduring the straight legge4 position of mid-stane. Itwould seem that the sane foot acrommodations takineplace at heel strike. and push-off in the running gaitwere also occurring in the racewalking git. During themid-stance phase of racewalking horuEver, the petvisdropped latenally, resulting in hip adduction and alateral shift in th€ center of gravity toward the stanseleg. An'increased medial ground reaction fofce urasnec€ssary to decelerate the lateral shift and to begin thestrift of the pelvis to the opposite sideforthenext stanceIeg, The reversal from medially to laterally directedground reaction forces occurrd at about 6AVo of racr.=walking stance.

The mean c€rter of pressure pattern during the com-petitive velocity of racewalking is displaysd in Figure4. After determining the actual location of the x and ycoordinates on the stroe outliqe for each subject, datawere normalized to shoe length, Figure 4represents thcaveqge location ofthe norrnalized data with each pointbeing separated by a percentage of average stance timeand thusthe distanc€ apan indicates the rate of change

.t

A BIOMEqHANICAL ANALYSS OF RACETI/ALKING GAIT

?at

\i

of position of the center of pressurc. The pathways ofthe aenler of pressure were also calculatod for therunning grut. Eight of the subjects displayed patrernstypical ofrearfioot strikerq and two displayed patterns:qrpi€l of mid-foot stril<ers m identified by Cavanaghand l-afortune (5), All subjec-ts made initial contactwith the ground more posteriorly durirrg racewatkingthan rutrning The pathrraV of the center of pressureprogessed anteriorly sliehtly modial to the mid-line ofthe shoe during approxinnately the first third of thestanc€ perioa. A rapid forward and lateral progressionoocurred during the uriddlE third of stancE, a timepeiiod corresponding to tfie strraight-fceged position andmaximal lateral deviation of the pelvis. During thelatt€r $tages of sqpporg the pathway of the ce.nter ofpressur€was similarto that ofrunning withfhe pathrrvay-returning to the mid-line:and the forcc fallipg belowthreshold levels prior to toe-off.

Figure 5 shows internal joint momentq nomralizedby ttodf weight and lower extremity length, plotted asa pe&€ntage of stance aud $ving time. The dorsiflexionmoment exerted during the early stance phase of race-ualking results from eccentric muscle contraction us6d

i\ mrr

45r

Rifl-

ilcE fitr- -Ja||

FGtlt

GoEtlI-

3oGtlOo.L

Figure }-Av.er.ge grou4Gfoot rcecticn focc etrveco! fO goqpetitive naaewalkers during three gdt.e-*,_ditiom exprcssed as multlple of body ryignt aldplotted ts. DqTeatrgF of sFqce tine.

GIF

J

ollI

to demleratc the downrryard motion of the tiody duringstance, a role asumed by the knee in ru-nm'ng Themagnirude of the plantar flcxion rnoment, averaged asa pel€nt4F of the gaitrltele, was about the same forall thr€e gaits but q€rt€d for a longer p€riod duringrunning

At the knee joint, the nrnging gait produc€d an€xtensof moment during the entire stanc€ phase whilethe $ance pbase of rmlking asd race*alking producedan initial extensor momenq a flexor moment, andanother extensor moment. The nrles of rawwalkingrquire that thc lnee be fully:extended fior some pgrioddurirrg the stance phase. An external hyp€rextensiotrmomert, a moment preletrt at knee angles greater thah175", was exerted at the knee for muc! of the stanc€phase during racewalking This external hl1errxtensionmoncnt rc$rltd in an internal fledon lromert whichoccuned at a time corr,g;ponding:closely tg the straight-legged position. This intenral flexion moment may beassumd to bc duc at least partiallyto posteriorcapsuleand ligament forces rather than flexor muscle for,ces.This enternal hypere;telrsion moment also ircr,easedsienilicatrtly with increases in the velocity of raceuffalk-

/r7='='a:--- '?----i '-="'---"- 'r"""'

ri\l:lil

Figu,c 4-Mcan oetrtcr of lrersure Dottcm duri[g onpdtl+c racerrtldng pith peree[t ol rrcirage strrce tine"

ing When rhe flexiou whicb {asqi€rted during externalhyperextension moment was removed from the calcu-lation of mearr knee flexion, there uas no significantdifrerence actocs gaits" Murray ct al. (14) recodedelestrornyographic activily ip the hamstringsduring thestan€ phaBc of racewalking which was explirined as adynamic ligamentous role of that group to pr€ventstretching of the posterior capsule.

Thc Beak knee extension moments at push-off(righttoc-off; for botb velocities of raceualking n'gre rignifi-eantly greater than walking or running The exteusionforcesat the knee seem ofgreaEr im@rtanae to achievepush-off in nacewalking whpreas running seems to relyupon p.lantar flexion furces as well.

Hip flexion and extension moments €xhibil similarpattems in all three gait modcs when plotted as apclceatage of cycle time. Each displays a short periodof extension moment followd by a longer period offlexion moment lasing into middte sojng whicir isfollowed by a second extension moment in late srping.Plotted across the rycle, hip abduction moments existedduring stan€ for:all gait modes with racewalking dis-

MEDICINE AND SCIENCE TN SPORTS AND qERCISE

lF rt rEE

--'+- lrtlo

-*ilfiSure s-AretaCa hternd &int Domenb df fO,competlrivc.noFra||rcrs dqiEg tbree gdi coodittm mrorlized by body wqigbt rdlorer ertreurity lilgth ud plotted rs r pereoqe of stance rDdswiry time.

,

playrng a highen magnitlde than walking 6; lqpningMuch of the abduction moment presmt duriag the€arly stance phase of raeualking nas a result of eccen-tric muscle contractions wtich control lateral pelvic tilt

The zun of all of the averagp moments may be usedas an estimate of total musclc activity during galt.RaCeualking produes a significantly higher total mo.metrt than the othcr gait modes. This would indicatetbat the faster velocity of racewalking had a higher rateof muscle activity compared to running at the sasespe€d and that individual muscle activitie were dif;er-ent for these wo gait modes. C-avqgna and Margaria (4)identified shifts from potential to kinetic energy asapproximately equal in magnitude io walking up tospeedS of 7 km'h-t. As ryeeds incrased up to 12 k6.h-r, oxygen consumptions were noted to b€ geat€rduring walking than running indicating mor€ activeshorteling of muscles in walking. Jt hes been shownthat the mechanics of normal and competitive walking

FIr,l

Ihl'i:a,

|'-!!Jllrta.,gtc.otaaa6LOIJ

liIT

I

i

I

rr^rcc__.{f- lrrro

A BIOMECHANICAL ANALYSIS OF RACEWALKING GAM

differ at spcgG.glwt€r than 7 km.h-!. Cav4gna andFranzefi {3) concluded that, at high speeds of race-watking, efficiency inereased with spd as a result ofthe storage of mechanical energy in muscle. Marchettiet al. (11), however, point out thal race*alking repre-sents a less e{frcienr form of looomotion than rulningsine the rnechanism of stored elastic enqr€y does notope,fate tO the same extent as in running. In the pfesentinvegigpliel, the significaatly higheraver-age tsial mo-ment present in racewalking at speds of 13 km.h-rsuggests greder energy utilization than runniqg at thesame speed.

In summary; it appears that the skill of raewalkinghas unique biomec-hanical properties which differen-tiatc it from the gits of walking and rrnning Theangulardifferencer in the motion ofthe lowerextrenritywhich exist between walking and raceqalking seemrccessary in order to &ttain inereased veloci-ties of gaitwithin the rules of racewalking and to modulate theexcursions ofthecenter of mass. The increasedanterior

REFERENCESI. ANDRTa€trr, T. P, J. A. OqLq and J. O. G^r-Ai{'rE \{alking

speed as a ksis fot normal and abnormal gait mcasuremeats J.Siomech. t6.26 t-268, 1977.

2. Blrng B. T- Functional variabilityof the lorvetexrrcmiryduringtle suppo$ phase of nrnning Me.d. Sci- Sports Exerc I11328'331. 1979.

3. Cr,vecxe, G. A. and P. FneNzE-rn. Mcchanics of competitionvalking J. Physiol. 3l 5:243-25 1, 1981,

4, Crv*cNl, G. A. and R. Mlnc*nle. M*banics of ualking /.Appl, Ph -vs iol 2l:,27 t 4 8, 19(f,.

5, CevrNrcn, P. R. and M" A. LAT0RTUNa Ground reacdon folocsin dista{ce nrnning. J. Eloueh. l3:397-4(X, 1980.

6. DotreLlss, B. L. and G. E, Gmns'_rr. Biqmcchanics of elitejunior race walkers- Irr $pans Biomrchanics, J. Tcrauds, lLBrthelq E. Kreighbaum, R. Illann, and J- Crakes {Eds"}, DelMal CA: .A,cademic PubliSers' 1984,pp..91-96.

7. FsFffoN, R. M. Racc $dkingground regction forces In: SpozsSiomechanics,J. Tereudc K. Barthels, E Kreighbaurn, R Mann,and J. Crakes (Edr). Del Mar" CA: Academic Publishers" 1984,pp.6l-70.

8. FRANxLTN, B. A., K P. K.lrulr-, T, W' Mon, and I{. lCHer r ri561". glaracteristics of national-clase racewalkers" Piys.Spoasmed- 9: l0 l-108. 1981.

9. Intgmatonal Amateur Athlefic Fedemtion Hoadbook Wesr Sus-ser, Englart* Mar$allarts Print Service, l9M,,p. 16Z-

10. M^NN, R. A,. and J. Fllcy. Sibrnechanics of walkiog running;and sprinting Am. J, Sports Med.8:345-:350, 1980.

I l- MARClrErrr, M,, A. Crnrozzo, F. Frcure, and Ii Fruo. Race-walking versrs ambulation and running ln: Biwiechanrcr, Yol-

453

and verticalgound reaction forpes of racewalkingoverwalking arc asociated with insreasd propulsive forcesof thc stance phasp in order to attain increased stridelength and v€locity. The incrcased medial @mponentof the ground teaction force ,prcsent during the mid-stance ph4s€ of racewalking appcm to be a compen,salory'foroc nmpqry to dewlerate and revcrse ttrclatcral pelvic shift in prcparation for the next staoceleg- The momentsofforu pr€$entatthe kneeand anklejoints during the stan@ phase of naceuaalkin€ appearneceps?ry in order to pErfilm thc joint meshanics pre-scriH by the rules of raoewalking.

This vrro?* rvas suFofted ln pffi ry a gntra frorn tp BoSFttMarathon Spcts Medicine Fund.

Re$rit address h'Ray G; B.rrlelt, Ph.D.: Progtr?rn in Phy:#alT|ler@V, Schod d l"leal$ Rdated russfr'ns, Urlversity of pitts-buqh, Pittsburgh, PA 1526r.

Address fur oorespsdencq Madtyn A. Gafra, Sc.D.. Deparrnentof Hdlh, Spc|t, and l-cislre Sfirdi{rs, Boston Bo$re @q dHunan Oevdopnefi Frortsslibn$, lbrt|easilqr'r ltdveGlity,,8oEton,MA 02115.

VIII-& fL Matsui aad IC Kobarahi (E<ls). Chammign,.ILHuman Kineticg l98a pp.669-6?5.

12. Msurn, D R. and L G, C. E Puctr" thc relation of oxlryenintake and wlocity of nalkipg and running in competition walk-u* J. Physiol. 197 :7 17 -721, 1968.

13. Munuv, M. P., ,{, B. DRouc}rr, and R C-'Ko*,v. Walkingpattenp ofnormal men.,l, Bow,Ioint Surg. 4G*33,r3&, 1964.

14. MuRnAy, M. P. G- Gurgu, L. lvtorlJrct& and G. Grs,DNE&Kiuemadc and eloqtrmyographic pafterns of olympic raccwalk.qts. Al L J. Spns Med. I l'68-74, 1983.

15. FerNe, A. tl. A comparison of ground reaction fomes in ncewalkins with thqe in normal walking and ruuting ltil. Eiottte..chanics, Vol. VI-A E" Asmussn a[d IC Jorgpnscn (Eds.). Balti-rnore, MDh Univsnity Park Prcss, 1978, p.293-3Q2.

16. PHtLLtps, S. J. and J. L. JB\6EL Kinemalies of race uatking. In:Spoas Biomeclwnia, J. Teraud+ IL Bsrrhcl* E. KreighbaunqR. Mann, aad-J. Crates (Eds.). Del Mar, Ctu Academic hcsq1984, pp. 7l-80.

17. SAUNDEnS, J. B- or C- M. V, T. INilAN, trDd H. D EsB.H^nr-The majordeterlninanrs in normal and ptholqgical,gliL J. Bonetoint Surg. 35-tr':543-J58, 1953"

I8. S|rroN, S. R., S. D Drursclr, R, M,'Nuzo, M, I Mllrsoun, J.L. Jrcrsox, M. Koxnrex, and R li RmgNrHi\u Genu reur-vatum in sFasric cerebral paby. t Bone loiru $urSl. 6OA:EE2-894, 1978.

19. SrrrHrru-eNo, D. H. and J. L Hesv- Measuremefi of gaitmcvErients from motion picturc film. l. Bone Joint sarg. 54.A:787-797.1972-