Critique of two research articles

A

O D S P r M R c C T ©

K

I

t T H t o n I b

p [ t o i t

0 d

Physiotherapy 97 (2011) 250–255

Effect of motion control running shoes compared with neutral shoes on tibial rotation during running

Alice Rose a,∗, Ivan Birch b, Raija Kuisma a a Division of Physiotherapy, School of Health Professions, University of Brighton, UK

b Faculty of Health and Human Sciences at Thames Valley University, UK

bstract

bjective To determine whether a motion control running shoe reduces tibial rotation in the transverse plane during treadmill running. esign An experimental study measuring tibial rotation in volunteer participants using a repeated measures design. etting Human Movement Laboratory, School of Health Professions, University of Brighton. articipants Twenty-four healthy participants were tested. The group comprised males and females with size 6, 7, 9 and 11 feet. The age

ange for participants was 19 to 31 years. ain outcome measures The total range of proximal tibial rotation was measured using the Codamotion 3-D Movement Analysis System. esults A one-tailed paired t-test indicated a statistically significant decrease in the total range of proximal tibial rotation when a motion

ontrol shoe was worn (mean difference 1.38◦, 95% confidence interval 0.03 to 2.73, P = 0.04).

onclusions There is a difference in tibial rotation in the transverse plane between a motion control running shoe and a neutral running shoe. he results from this study have implications for the use of supportive running shoes as a form of injury prevention. 2010 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved.

m p s t f a w r t I s [

a a

eywords: Tibial rotation; Running; Motion control shoe

ntroduction

In today’s society, one of the greatest challenges faced by he Department of Health is the growing epidemic of obesity. he cost of overweight and obese individuals to the National ealth Service is estimated to be £4.2 billion [1]. This has led

o considerable media attention on the wider health risks of ur sedentary lifestyles, and has resulted in an increase in the umber of people participating in recreational running [2]. n order to understand the effect of running shoes on gait, a asic knowledge of lower limb mechanics is essential.

Within the first 20% of the stance phase, the subtalar joint ronates to allow solid contact of the foot with the ground 3]. As forward progression continues through the middle of he stance phase, maximum pronation and ankle dorsiflexion

ccur. Pronation is a normal part of the running cycle because t allows for shock absorption and accommodation on uneven errain [3]. However, in some individuals, excessive pronation

∗ Corresponding author. E-mail address: [email protected] (A. Rose).

t i h f a f

031-9406/$ – see front matter © 2010 Chartered Society of Physiotherapy. Publis oi:10.1016/j.physio.2010.08.013

ay occur for various biomechanical reasons [4]. Excessive ronators present with a broad range of pathologies, such as tress fractures, achilles tendonitis and iliotibial band (ITB) endonitis [4]. Yates and White studied naval recruits and ound that those with a pronated foot type were almost twice s likely to develop medial tibial stress syndrome compared ith those with a normal or supinated foot posture [5]. A

isk estimate revealed that recruits with a more pronated foot ype had a higher relative risk (1.70) than injury-free recruits. n an attempt to minimise the risk of injury, athletes have tarted to seek specific equipment, particularly running shoes 5].

Neutral cushion shoes are generally best for runners with n excessive supinatory gait to provide additional shock bsorption, whereas motion control shoes are better for he moderate to severe overpronator. Motion control shoes nclude a reinforced heel counter and a denser midsole to

elp control any excessive pronation [5]. Clarke et al. [6] ound that shoes with a positive heel flare and a hard midsole llowed significantly less maximum pronation and total rear- oot movement compared with shoes with a softer midsole.

hed by Elsevier Ltd. All rights reserved.

therapy

a o a s n M d p i i i fi t c w s m [ i c n I m m

i r a D a r i h i a C T c T d n C e

i r i w i

M

t

r m i

P

m v s i t M t

s 1 h d o

I

I b l L u i p w

P

p n t

t o i b X e c

• • •

A. Rose et al. / Physio

Research has shown that motion control shoes also have n effect on tibial internal rotation [7]. This is a normal part f the gait cycle and occurs in conjunction with pronation nd hindfoot eversion due to the mitered hinge effect of the ubtalar joint [3]. Running can lead to an increase in inter- al tibial rotation for people with excessively pronated feet. iller et al. [8] found that runners with a history of ITB syn-

rome demonstrated increased tibial internal rotation. They redicted that an increase in internal rotation of the tibia could ncrease the strain in the ITB, and therefore be a contribut- ng factor to ITB syndrome. However, the critical degree of nternal tibial rotation leading to injuries has not been con- rmed, and research is required in this area. It is reasonable

o assume that by reducing rearfoot pronation with motion ontrol shoes, tibial rotation would consequently decrease hen running. This may support the use of supportive running

hoes as a form of injury prevention. To date, there is mini- al evidence on the effects of running shoes on tibial rotation

8–10]. Stacoff et al. [10] used intracortical bone pins with njury-free participants to measure the effects of shoe sole onstruction on skeletal motion during running. They found o statistically significant change in tibiocalcaneal rotations. t was concluded that the tibiocalcaneal kinematics of running ay be unique to the individual, and shoe sole modifications ay not be able to make substantial changes. Running gait can be analysed using a number of methods

ncluding real-time observational gait analysis, high- esolution cameras and video-recording devices, force plates nd computer systems. Computerised three-dimensional (3- ) motion analysis measurements are currently a widespread

nd useful tool for both clinical practice and biomechanical esearch. The Coda motion 3-D Movement Analysis System s able to measure locations of active markers in 3-D with igh resolution and accuracy. Maynard et al. [11] studied the ntra- and inter-rater reliability of gait measurements using Cartesian optoelectronic dynamic anthropometer (CODA, harnwood Dynamics Limited based in Leicestershire). hey suggested that natural variation of the participant’s gait ycle may be overcome by capturing at least three gait trials. hey concluded that they had not shown complete repro- ucibility of gait measurements using CODA, but this does ot suggest that CODA is unreliable. Many studies have used ODA in the past and none have recorded any anomalies or rrors attributal to CODA itself [12,13].

Published literature suggests that running footwear can nfluence lower extremity kinematics and kinetics. Currently, esearch into the effects of running footwear on tibial rotation s limited. Therefore, the aim of this study was to determine hether motion control running shoes reduce tibial rotation

n the transverse plane during treadmill running.

ethods

An experimental study was conducted with two condi- ions: neutral shoes and motion control shoes. The effect on

o o I

97 (2011) 250–255 251

ange of tibial rotation in the transverse plane during tread- ill running in the two styles of running shoes was measured

n degrees.

articipants

Ethical approval was gained from the School Ethics Com- ittee at the University of Brighton. Thirty-two consecutive

olunteers from the University of Brighton health profes- ional courses who responded to a recruitment e-mail were ncluded. This e-mail included an information sheet for par- icipants to read before volunteering. On arrival at the Human

ovement Laboratory, all details of the study were explained o participants and they were asked to sign a consent form.

In order to be included, participants were required to have ize 6, 7, 9 or 11 feet and be able to run comfortably for 5 to 0 minutes. Participants were excluded from the study if they ad a history of cardiovascular problems, a lower limb con- ition that was exacerbated by running, a vestibular disorder r an allergy to hypoallergenic adhesive tape.

nstrumentation

A Biodex RTM 500 treadmill (Biodex Medical Systems, nc., New York, USA) was used for all participants and for oth conditions. The treadmill was placed in the middle of the aboratory with one CODA MPX30 (Charnwood Dynamics td., Leicester, UK) scanner unit in front and one scanner nit behind. These units picked up signals emitted by the nfra-red ‘active’ markers which provided an immediate and recise 3-D measurement. All information from the scanners as stored on the CODA computer system.

rocedure

A pilot study was carried out on one individual to test the roposed methodology. No problems were experienced, so o changes were made to the methodology. The data from he pilot study were included in the main results.

The laboratory was set up before participants arrived on he first day. The origin or zero point of the system’s mutually rthogonal measurement framework was established by plac- ng a single marker in the centre of the treadmill, mid-way etween the two sensor units. The alignment of the system’s , Y and Z axes was set using two additional pairs of mark- rs. The orientation of the three axes used throughout the data ollection was as follows:

positive Y axis – direction of walking; positive X axis – to the participant’s right; positive Z axis – up.

All data were collected using the right leg, and the axis rientation was set in accordance with the recommendations f the Standardization and Terminology Committee of the nternational Society of Biomechanics [14].

252 A. Rose et al. / Physiotherapy 97 (2011) 250–255

F

i c n A g l i m C ( u i a T p t c f a m i o [

a w s c t a o w t t t M w s

F

Data analysis

The first useable set of data of the three sets collected per participant was used for data analysis. The average range of

ig. 1. Neutral shoe: Mizuno Wave Rider.

Participants were asked to attend a single session last- ng approximately 30 minutes. Participants were asked to ome wearing shorts and to put on the neutral pair of run- ing shoes (Fig. 1) before marker attachment took place. ctive markers were attached using double-sided hypoaller- enic adhesive tape to the following landmarks on the right eg: medial condyle of tibia, head of fibula, tibial tuberos- ty, two-thirds of tibial shaft, medial malleolus and lateral

alleolus (Fig. 2). Marker placement complied with the Joint o-ordinate System recommendation [9]. Technical markers

medial condyle of tibia and two-thirds of tibial shaft) were sed to define the co-ordination of the virtual markers dur- ng running trials. Batteries were connected to the markers nd attached to the leg using hypoallergenic adhesive tape. he safety clip of the treadmill was attached to the partici- ants and they were given a 5-minute practice run to allow hem to adapt to the treadmill motion and the shoes. Lav- anska et al. found that it takes 5 to 6 minutes to become amiliar with treadmill running; therefore, participants were cclimatised for this period prior to commencement of easurements. The treadmill speed was then gradually

ncreased to 8 km/hour [15]. The average running speed f unimpaired young adults is between 7 and 9 km/hour 15].

Participants were shown the emergency stop button and ssured that they could stop running at any time. Confirmation as gained that the participant was happy running at this

peed. On confirmation, lights were switched off to do a ‘test’ ollection of data to ensure that all markers were in view of he scanners. One researcher carried out all the tests with an ssistant to help. If all the markers were in view, three sets f data were collected and saved to the computer. If a marker ent out of view at any point, the treadmill was stopped and

he markers were adjusted accordingly or retaped. Once the hree sets of data were saved, participants dismounted the readmill and changed into the motion control shoes (Fig. 3).

arkers were checked for secure attachment. The procedure

as then repeated for the motion control shoes. All data were

aved to a password-secured computer. F

ig. 2. Front view of active markers.

ig. 3. Motion control shoe: Mizuno Wave Inspire.

A. Rose et al. / Physiotherapy

Table 1 Total range of proximal tibial rotation in the XY plane of the Cartesian optoelectronic dynamic anthropometer for the two running shoes.

Participant Total range of tibial rotation (degrees)

Neutral shoe

Support shoe

Difference (◦)

1 10 16 −6 2 4 9 −5 3 5 9 −4 4 10 14 −4 5 12 14 −2 6 7 7 0 7 6 6 0 8 22 21 1 9 17 16 1 10 8 7 1 11 5 4 1 12 9 8 1 13 15 14 1 14 7 5 2 15 23 21 2 16 9 6 3 17 18 15 3 18 18 15 3 19 10 7 3 20 12 8 4 21 11 6 5 22 20 14 6 23 10 4 6 24 28 17 11 M

m p c t C w t s

R

p e m u i T

f d P d t c

t ( w t m 1 t

r t

D

c p p s d s m r w e c

d t s h a b w r p s t s i t 1

b n s i k s t n reduction in peak rearfoot eversion for runners in the motion

ean Standard deviation 12.3 (6.4) 10.9(5.3) 1.4 (3.8)

otion of the leg in the transverse plane during the contact hase of running was calculated using CODA. The data were opied into Microsoft Excel, and graphs were plotted to show he total range of proximal tibial rotation in the XY plane of ODA for the two running shoes. A one-tailed paired t-test as used to test for differences in tibial rotation recorded for

he two experimental conditions (neutral and motion control hoes), and the P-level was set at 0.05.

esults

All 32 participants took part in the study, but only 24 articipants provided adequate data for analysis. Naturally, very individual’s running gait is different and sometimes this eant that the signals from the active markers were not picked

p by the scanner units. This occurred for eight of the partic- pants; therefore, the data were excluded from the analysis. able 1 shows the data collected from the 24 participants.

The paired t-test revealed a statistically significant dif- erence in tibial rotation between the two conditions (mean ifference 1.38◦, 95% confidence interval 0.03 to 2.73, = 0.04). This showed that there is a statistically significant

ifference in tibial rotation in the transverse plane during readmill running when using motion control running shoes ompared with neutral running shoes.

c p r

97 (2011) 250–255 253

The overall trend showed that the total range of proximal ibial rotation for participants running in motion control shoes mean 10.9◦, standard deviation 5.3◦) was smaller compared ith running in neutral shoes (mean 12.9◦, standard devia-

ion 6.4◦). The largest individual difference was 11◦, but the ean difference between the two conditions was small (mean

.38◦). The tibial rotation in five participants was smaller in he neutral running shoes (Table 1).

The largest difference was 11◦; two participants had a total ange of tibial rotation of 10◦ in the neutral shoes and 4◦ in he motion control shoes.

iscussion

The aim of this study was to determine whether motion ontrol running shoes reduce tibial rotation in the transverse lane during treadmill running, with implications for injury revention. The findings suggest that there is a statistically ignificant difference in tibial rotation in the transverse plane uring treadmill running between motion control running hoes and neutral running shoes. The total range of proxi- al tibial rotation was generally reduced when participants

an in motion control shoes. Although the mean difference as rather small, some participants demonstrated a differ-

nce which was half of the total range, and hence could be linically significant in their cases.

Clarke et al. [6] analysed the effects of different shoe esign parameters on rearfoot control in running. They found hat participants displayed an average of approximately 2.7◦ maller maximum pronation when running in shoes with a ard midsole compared with a soft midsole. If rearfoot motion nd tibial rotation are taken as a coupled mechanism [3], it can e assumed that this is the mechanism by which tibial rotation as reduced in this study. Controlling rearfoot motion with

unning in motion control shoes will subsequently reduce the articipant’s tibial rotation. The results from the current study upport this assumption because the total range of proximal ibial rotation was reduced when running in motion control hoes compared with running in neutral shoes for the major- ty of participants (79%). For example, one participant had a otal range of tibial rotation of 28◦ in the neutral shoes and 7◦ in the motion control shoes (reduction 11◦).

The results from the current study relate to similar research y Butler et al. [7], which showed that in low arched run- ers, peak tibial internal rotation decreased in motion control hoes and was increased in cushion shoes. They found no nteraction between high arched runners and lower extremity inematics. Since arch height was not measured in the current tudy, it is unclear if this factor contributed to the degree of ibial rotation. To confirm this association, further research eeds to be undertaken in the area. Butler et al. [7] found no

ontrol shoes. This is unusual because if there was a cou- led relationship between rearfoot eversion and tibial internal otation, it would be expected that both of these movements

2 therapy

w u i o

e p t s s b t S k b t w p l s S b m r i p

t n [ a m b m m e i a m s a a a

b n c g a t p

o m f t

a m A d t e t t f

t t s h e t e p r T p a

b n c

u f c f p t t t t

C

t w r d m h r s r

54 A. Rose et al. / Physio

ould decrease when running in motion control shoes. These nusual results may be due to the motion control shoes having ncreased support in the midfoot; therefore, the positioning f this support would not have influenced rearfoot motion.

This study does not provide direct support for the hypoth- sis that motion control running shoes are a form of injury revention; however, it does have some supporting implica- ions. Miller et al. [8] found that runners with a history of ITB yndrome demonstrated more tibial internal rotation, thus uggesting that an increase in tibial internal rotation could e a contributing factor to ITB syndrome. The links between ibial rotation and running injuries are further supported by tergiou and Bates [16]. They investigated knee and ankle inematics in five runners, and showed a strong relationship etween pronation and knee joint function via tibial rotation; his was identified as a possible mechanism for injury. They ent on to explain that overpronation could lead to maximum ronation (and subsequent internal tibial rotation) occurring ater in the stance phase. This could then lead to soft tissue tress around the knee or patellofemoral malalignment [16]. ince anyone with a lower limb condition that is exacerbated y running was excluded from this study, the hypothesis that otion control shoes could be a form of injury prevention for

unners was not tested. However, the links between increased nternal tibial rotation and running injuries found to date seem romising, and there is scope for further research.

The Codamotion 3-D Movement Analysis System is easy o use, and previous studies that have used CODA have ot found any anomalies or errors attributal to CODA itself 12,13]. However, there are certain limitations that may have ffected the outcome of the results. The issue of skin move- ent artefacts when using skin markers in gait analysis has

een raised [3], although tibial rotation is considered to be ore accurate than other measurements of knee and thigh ovement [12]. When comparing bone pins with skin mark-

rs, Reinschmidt et al. [12] found an average error of 1.1◦, mplying that tibial rotation can be determined with reason- ble accuracy using skin markers. In this study, the active arkers were placed over bony structures where underlying

oft tissue was minimal, with the hope that skin movement rtefacts would be minimised. This is especially the case round the ankle where the skin is tightly bound and there is decreased chance of soft tissue and skin movement.

Maynard et al. [11] studied the intra- and inter-rater relia- ility of gait measurements using CODA, and suggested that atural variation of the participant’s gait cycle may be over- ome by capturing at least three gait trials. In this study, three ait trials were captured for each participant and the first use- ble set of data was used for data analysis. This improved he reliability of the measurement technique and minimised otential error.

Although treadmill running allows for easy, continuous

bservation and monitoring, it may cause variation in move- ent pattern compared with overground running. Therefore,

or a true representation of running biomechanics, tibial rota- ion would need to be measured over ground.

A

F

97 (2011) 250–255

Five participants did not follow the trend of results nd showed an increase in tibial rotation when wearing otion control running shoes compared with neutral shoes. number of factors could have contributed to this unusual

ifference. During data collection, it became apparent that here was inconsistency with the level of treadmill experi- nce between the participants. This could have resulted in he less experienced participants becoming fatigued by the ime they reached the second test condition. Links between atigue and rearfoot kinematics have been recorded [7,17].

Participants in this study ran first in the neutral shoes and hen in the motion control shoes. This may have contributed o some of the increases in tibial rotation in the motion control hoes. To minimise possible order effects, the shoes should ave been presented in a randomised order. This would have liminated a potential confounding variable and strengthened he internal validity of the study. Due to limited funding, only ight pairs of shoes were used for the study, which excluded articipants outside the available shoe sizes. Therefore, the esults may not be representative of the general population. his study would need to be reproduced with a larger sam- le size (including all shoe sizes) to make the results more pplicable to the wider population.

The results from the study are open to observer bias ecause the researcher collected the study data. To elimi- ate this potential confounding variable, an independent data ollector should have been used.

A number of ideas for further research arose while eval- ating the current study. As highlighted above, the results rom this study have implications for the theory that motion ontrol running shoes could be a form of injury prevention or runners. The experiment could be expanded to include articipants with running-related injuries. Researchers could hen investigate to see if injured participants displayed more ibial rotation in the neutral shoes, and furthermore if this ibial rotation decreased for these participants when wearing he motion control shoes.

onclusions

The results from this study show that there is a difference in ibial rotation in the transverse plane during treadmill running hen comparing motion control running shoes with neutral

unning shoes. This is indicated by the statistically significant ecrease in the total range of proximal tibial rotation when otion control shoes were worn. Previous research findings

ave demonstrated links between increased tibial rotation and unning injuries [8,9]. The results from this study therefore uggest that supportive running shoes may have an important ole to play in running injury prevention.

cknowledgements

The authors would like to thank Amy Grimadell, Leigh agan and Nicole Nielsen for assisting with data collection;

therapy

P f B

E E

C

R

[

[

[

[

[

[

[

1997;6:177–85. [17] Gerlach K, White S, Burton H, Dorn J, Leddy J, Horvath P. Kinetic

A. Rose et al. / Physio

aul Stevens at Mizuno Ltd. for providing the running shoes or this study; and all the participants from the University of righton.

thical approval: School of Health Professions Research thics and Governance Panel at the University of Brighton.

onflict of interest: None declared.

eferences

[1] Department of Health. Obesity. Department of Health, Lon- don, 2007. Available at: http://www.dh.gov.uk/en/Publichealth/ Healthimprovement/Obesity/DH 6585 (last accessed 14.04.2009).

[2] Buist I, Bredeweg S, Bessem B, Van Mechelen W, Lemmink K, Dier- cks R. Incidence and risk factors of running-related injuries during preparation for a four-mile recreational running event. Br J Sports Med 2008;5:1–12.

[3] Dungan S, Bhat K. Biomechanics and analysis of running gait. Phys Med Rehabil Clin N Am 2005;16:603–21.

[4] Yamashita M. Evaluation and selection of shoe wear and orthoses for the runner. Phys Med Rehabil Clin N Am 2005;16:801–29.

[5] Yates B, White S. The incidence and risk factors in the development of medial tibial stress syndrome among naval recruits. Am J Sports Med 2004;32:772–80.

[6] Clarke T, Frederick E, Hamill C. The effects of shoe design parameters

on rearfoot control in running. Med Sci Sport Exerc 1983;15:376–81.

[7] Butler R, Hamill J, Davis I. Effect of footwear on high and low arched runners’ mechanics during a prolonged run. Gait Posture 2007;26:219–25.

Available online at www.s

97 (2011) 250–255 255

[8] Miller R, Lowry J, Meardon S, Gillette J. Lower extremity mechan- ics of iliotibial band syndrome during an exhaustive run. Gait Posture 2007;26:407–13.

[9] Eslami M, Begon M, Farahpour N, Allard P. Forefoot-rearfoot coupling patterns and tibial internal rotation during stance phase of barefoot versus shod running. Clin Biomech 2007;22:74–80.

10] Stacoff A, Reinschmidt C, Nigg B, Van den Bogert A, Lundberg A, Denoth J, et al. Effects of foot orthoses on skeletal motion during running. Clin Biomech 2000;15:54–64.

11] Maynard V, Bakheit A, Oldham J, Freeman J. Intra-rater and inter-rater reliability of gait measurements with CODA mpx30 motion analysis system. Gait Posture 2003;17:59–67.

12] Reinschmidt C, Van den Bogert A, Nigg B, Lundberg A, Murphy N. Effect of skin movement on the analysis of skeletal knee joint motion during running. J Biomech 1997;30:729–32.

13] Monaghan K, Delahunt E, Caulfield B. Ankle function during gait in patients with chronic ankle instability compared to controls. Clin Biomech 2006;21:168–74.

14] Wu G, Siegler S, Allard P, Kirtley C, Leardini A, Rosenbaum D, et al. ISB recommendation on definitions of joint coordinate system of vari- ous joints for the reporting of human joint motion – part 1: ankle, hip, and spine. J Biomech 2002;35:543–8.

15] Lavcanska V, Taylor N, Schache A. Familiarization to treadmill running in young unimpaired adults. Hum Mov Sci 2005;24:544–57.

16] Stergiou N, Bates B. The relationship between subtalar and knee joint function as a possible mechanism for running injuries. Gait Posture

changes with fatigue and relationship to injury in female runners. Med Sci Sport Exerc 2005;37:657–63.

ciencedirect.com