Background – Normal functioning of the lower limb depends on correct functioning of the foot. It is hypothesized that abnormal foot biomechanics associated with different foot types, may lead to abnormal stresses on proximal muscular structures. These abnormal stresses may eventually result in musculoskeletal injuries.
Method – This experimental pilot study investigated tibialis anterior (TA) muscle electromyography (EMG) activity during stance phase of gait in healthy participants with supinated (n=8), normal (n=10) and pronated (n=10) feet. Subjects walked on a gait analysis treadmill and EMG activity of TA was recorded simultaneously. The total activity of TA during four phases of stance was compared between the three groups.
Results – No statistically significant differences were found in the EMG activity of TA during any of the stance phases of gait for subjects classified in supinated, normal and pronated groups (p>0.05).
Implications – There is no evidence to support the view that foot type will lead to lower limb injury as a consequence of altered TA muscle activity.
Conclusion – It seems that foot type may not be a factor in the development of TA related overuse injuries. However investigation of more severe groups of pronated and supinated subjects may be more revealing.
Normal functioning of the lower limb during walking depends on correct functioning of the foot. The foot is the first part of the lower extremity contacting the ground. It is subject to ground reaction forces (GRF) as the heel strikes the supporting surface and this force is transferred proximally to the rest of the lower kinetic chain. Therefore, correct biomechanics and kinematics of the foot should allow the lower extremity to cope with GRFs during weight bearing (Donatelli 1985).
During the stance phase of gait the foot undergoes mechanical changes alternating between pronated and supinated positions. The foot supinates slightly at initial contact, pronates when it is in full contact with the ground and supinates again before toe off (Whittle 2002). The pronation element results from unlocking of the subtalar joint producing a flexible foot which allows the foot to adapt to the ground, bear weight and maintain stability of the body over the foot. The supination element at the end of stance phase leads to locking of the subtalar joint which helps to transform the foot into a rigid lever. Moreover, pronation and supination movements during stance phase assist in the distribution of normal forces through the lower limb kinetic chain (Donatelli 1985). The extrinsic muscles of the lower leg, including tibialis posterior (TA) and peroneus longus contribute to propel the body forward during gait as the foot alternates between these pronated and supinated positions (Donatelli 1985, Menz et al 2005).
Any disturbance in the alignment of the foot is thought to interfere with the mobility of the joints of the foot resulting in abnormal biomechanics and function of the foot (Donatelli 1987). This might lead to abnormal stresses being imposed on the structures in the lower leg and may cause injury (Neely 1998). Previous research has demonstrated a relationship between foot type and musculoskeletal injuries e.g. shinsplints (De Lacerda 1980), tibial stress injuries (Beck 1998), overuse injuries (Kaufman et al. 1999), tendonitis and fasciitis (Donatelli 1987). Extreme postures of supination and pronation may limit the ability of the foot to adjust to the ground (Cote et al 2005) causing increased demand on the surrounding musculoskeletal structures which compensate to keep the body stable (Hunt et al 2000). This may create abnormal tensile stresses on the extrinsic foot muscles including TA, tibialis posterior, peroneal muscles, gastrocnemius, soleus, and flexor digitorum and hallucis muscles (O’Connor and Hawill 2004). Myofascial inflammation, edema, myotatic contraction (a reflex contraction that occurs due to stimulation of the stretch receptors in the muscle), subperiosteal avulsions and pain may follow (De Lacerda 1980).
Furthermore, in supinated feet, there is a smaller contact area between the foot and the ground. Therefore there is potentially less sensory feedback from the foot to the central nervous system to support stabilizing function, consequently the demand on surrounding muscular structures may increase to compensate and maintain balance (Cote et al. 2005). In pronated feet, the stretch is increased on the invertor muscles including tibialis anterior and posterior which control eversion movement (O’Connor and Hawill 2004). This might lead to increased demand on these muscles to control eversion during walking.
During gait, TA maintains dorsiflexion of the ankle in preparation for heel strike (Whittle 2002) and by contracting eccentrically controls the lowering of the foot after initial contact (Nilsson et al. 1985). As an invertor of the foot TA can help control eversion during stance phase (Cornwall and McPoil 1994). TA is well placed to oppose the passive pronation movement (combined dorsiflexion, eversion and abduction) that tends to occur at heel strike (Donatelli 1985) and stance phase.
Several studies have investigated the effect of flat feet on lower leg extrinsic muscles activity (Hunt and Smith 2004, Gray and Basmajian 1968); however to date no studies have been found which investigate the effect of supinated type of foot. The aim of this pilot study was to investigate whether foot type affects TA muscle EMG activity during stance phase of gait. It was hypothesized that TA muscle EMG activity would be greater in pronated and supinated foot types compared with normal foot type. Sustained, increased, EMG activity in TA might act as a precursor for developing lower leg injury such as shin splints.
Twenty eight subjects, 19 females and 9 males, with no previous foot injuries, fixed foot deformities or biomechanical abnormalities participated in this study. They were classified into three groups according to their foot type; normal (n=10), pronated (n=10), and supinated (n=8). Ethical approval was granted by Sheffield Hallam University Ethics Committee. Prior to commencing the study all subjects read the participant’s information sheet and signed the consent form.
Classification of foot type
The foot type of the subjects left foot was classified as pronated, normal or supinated by a navicular drop test (NDT) described by Brody (1982). Navicular drop corresponds to the vertical movement of the navicular bone between subtalar neutral position and relaxed weight bearing position (Snook 2001). The distance between the tip of the navicular bone and the ground (navicular height, NH) was measured using a caliper (Draper 0-300mm Vernier Caliper) (see Figure 1) first from a sitting position and then from a relaxed standing position, with the subtalar joint in neutral position. The subtalar neutral position was determined by moving the foot alternately in inversion and eversion until equal depressions were palpated anterior to medial and lateral malleoli. The difference between the two measurements was calculated and accordingly the foot was classified as normal (6-9 mm), pronated (>9 mm), or supinated (<6 mm) (Loudon et al 1996).
The NDT is considered a clinically relevant tool for classifying foot type (Sell et al. 1994) which is both valid and reliable. Williams and McClay (2000) found high levels of validity of NH measures in weight bearing positions which is the basic measurement in NDT. This test was found to have good to excellent inter-tester and intra-tester reliability (Trimple et al. 2002, Williams and McClay 2000, Sell et al. 1994). Specific procedures were followed in this study to increase the reliability of the NDT; using a caliper to measure navicular height (Razeghi and Batt 2002) and calculating the average of 3 repetitions of the measurements (Sell et al. 1994).
Tibialis anterior muscle activity-EMG recordings during stance phase of gait
Tibialis anterior muscle activity was measured using a 2-channel handheld surface EMG system (Bagnoli, DELSYS Inc. Boston, USA). The recording 2 bar type electrode (DE-2.1, 19.8 mm x 5.4 mm x 35 mm) was attached to the skin using a Delsys Electrode Interface (made from medical grade adhesive specifically designed for dermatological applications). The placement of the electrode was one-third of the way along the line between the tip of the fibula and the tip of the medial malleolus (according to recommendations for sensor locations on individual muscles by SENIAM project, www.seniam.org). The two parallel silver bars within the electrode lay perpendicular to the muscle fibers. The reference electrode, a disposable adhesive conductive electrode disk (DELSYS) was placed over the lateral malleolus. Figure 2 shows electrode placement during the trial. The skin under the electrodes was cleaned with isopropyl wipes and dried well before the application of the adhesive interface. Hair at the area was shaved using safety razors and dry skin cells were stripped using surgical tape. Standardized procedures of skin preparation were undertaken to ensure better fixation of the electrodes and minimize any motion artifact that may interfere with the EMG signal (as recommended in Bagnoli EMG system user manual, by DELSYS). To ensure stability of the recording electrode, it was fixed with crepe bandage.
Fig. 1: Figure 1
Measurement of navicular height using a calliper.
Fig. 2: Figure 2
Electrode placement and EMG recording procedure on the gait analysis treadmill
The cables connecting the electrodes to the EMG amplification unit were fixed with surgical tape on the subject’s skin to minimize artifacts caused by dangling cables which may corrupt the recorded signals.
EMG recording was done while the subject walked on a gait analysis treadmill integrated with built-in force platforms (H-P COSMOS KISTLER Gaitway Treadmill type 9810AS10) (see figure 2). Subjects walked on the treadmill for 30 seconds with zero inclination and at age and gender-dependant average speed; (20-59 years: 1.4 m.s-1 for males and 1.3 m.s-1 for females) (Waters et al 1988). Vertical GRF data for stance phase of gait was collected using KISTLER Gaitway Software (version 2.04).
Recording muscle electrical activity using SEMG during gait demonstrates repeatability, reliability and consistency (Bogey et al 2003). Treadmills integrated with force platforms were previously found to be accurate for measuring joint GRF (Kram et al 1998). Quantitative and qualitative characteristics and overall pattern of gait on a treadmill were found to be very similar to walking over ground (Riley et al 2007), despite slight difference in stance duration (Warabi et al 2005) and muscle activation (Lee and Hidler 2008). Gait analysis treadmill was used in this study to facilitate simultaneous EMG recording as the leads cannot accommodate over ground walking.
Other variables such as gender, age, weight and height were measured at baseline since these were considered to be potential confounders of variations in foot type.
Processing and analysis of data
EMG signals were amplified with a gain of 1000 and digitized by a 16-bit analogue to digital converter. EMG and GRF data were then synchronized and analyzed using matlab program version 126.96.36.199 (Release 12). EMG data was full-wave rectified and band-pass filtered between 20-480 Hz with a sampling frequency of 512 Hz. GRF data was filtered with a low pass fourth order bypass filter with a frequency of 50 Hz and a zero lag. Four phases of stance were identified from the GRF graphs as recommended by KISTLER Inc (Figure 3) and the average muscle EMG activity during each phase was calculated from five steps of the left foot for each participant. Corrupted data which could not be processed by the algorithm of the modified matlab program was excluded and then for each subject five steps which had stance duration closest to the average stance phase duration were selected.
Fig. 3: Figure 3
Phases of stance phase used in EMG analysis as recommended by Kistler Inc. a= phase 1, b= phase 2, c= phase 3, d= phase 4
Statistical analysis of data was done using the Statistical Program for Social Sciences (SPSS version 15.0) for Windows, the Shapiro-Wilk test was used to check for normal distribution of the data (Field 2005). Since data was not normally distributed, the Kruskal-Wallis test was used to compare average muscle EMG activity in each sub-phase of stance phase between the three groups (normal, supinated and pronated) (Field 2005).
Baseline characteristics of the 28 participants (19 females and 9 males) are shown in Table 1. There were no significant differences found between the three groups regarding age, height or weight (P>0.05). As expected there was a significant difference in ND (p<0.05) (Table 1).
Table 1: Navicular drop, age height and weight (means, standard deviations) for the three comparison groups
NavicularDrop (ND) (mm)
Navicular drop, age height and weight (means, standard deviations) for the three comparison groups
Tibialis anterior muscle activity
Table 2 shows the median values and ranges of tibialis anterior EMG activity for the three groups during the four sub-phases. During the four phases of stance phase, there were no significant differences in tibialis anterior muscle EMG activity between the three comparison groups (p>0.05) (Table 2, Figure 4). There was no correlation between the total muscle EMG activity during stance phase and baseline variables including ND (r=0.046, p=0.816), age (r=0.024, p=0.902) and weight (r=0.163, p=0.408).
Table 2 - Median values, ranges and P-values of tibialis anterior muscle EMG activity (V) for the three groups during the four phases of stance phase
Table 2 -
Median values, ranges and P-values of tibialis anterior muscle EMG activity (V) for the three groups during the four phases of stance phase
Fig. 4: Figure 4
Box plots showing differences in TA EMG activity between supinated, normal and pronated groups during the four phases of stance.
The results of this study revealed no significant differences in the EMG activity of TA during stance phase of gait between the three foot types in asymptomatic subjects. Several previous studies have investigated the activity of selected extrinsic muscles in the lower leg during walking in subjects with flat feet compared to normal subjects. Hunt and Smith (2004) found a relative increase in TA muscle EMG activity at heel strike in flat-footed patients. Gray and Basmajian (1968) found continuous activity of TA during foot-flat stage of stance phase in subjects with flat feet which was not detected in normal subjects. In normal subjects TA was silent. They attributed this finding in flat feet to the possibility that TA muscle continues its action in order to maintain foot inversion to allocate the weight on the lateral border of the foot rather than the medial border. This finding corresponds with Hunt and Smith’s (2004) observations that the greater invertor moment in the flat-footed group during this same stage of stance (foot-flat) indicates a more forceful eccentric contraction of invertor muscles including TA.
The results of the current study could not be easily compared with these studies as the methodology was different and analysis of EMG during stance phase was dissimilar. In both studies sample inclusion criteria was not based on clinically reliable or accurate measurements of pronated or flat feet e.g. Hunt and Smith (2004) depended on clinician’s referral, whilst Gray and Basmajian (1968) relied on observation of medial longitudinal arch to classify flat foot. The visual inspection method used for foot classification has been shown to be highly variable and unreliable (Cowan et al 1994). Depending on observation might lead us to hypothesise that in both studies excessively pronated feet, which were easily classified by an observation method, were more likely to be included and could have explained the significant differences in tibialis anterior muscle EMG activity found between the comparison groups at certain stages of stance phase.
This study adds to previous research as the effect of supinated foot type on tibialis anterior muscle EMG activity was included in the study. As far as we are aware this has not previously been investigated.
Study strength and limitations
In the current study eight participants in the pronated group had degrees of ND which could be regarded as close to normal (range 10.0 to 12.3 mm). Two participants had ND of 17.0 and 20.3 mm which could be considered extreme pronation (ND>15mm) (Brody 1982). The ND in the supination group was close to normal in 4 subjects (range 4.0 to 4.6 mm) and might be considered relatively extreme in only 4 subjects (range 0.5 to 0.56 mm). This small number of subjects having more extreme levels of supinated or pronated feet might not have allowed for significant findings to be shown. A greater number of subjects with more severe levels of pronation and supination might have allowed detection of differences between groups.
A gold standard method of classifying foot type is not available. In the current study, the NDT was chosen to classify foot type as it is considered a reliable and valid method of foot classification (Williams and McClay 2000) and it has been widely accepted and used in previous research (Razeghi and Batt 2002). Using the NDT overcomes the limitations identified in previous studies (Hunt and Smith 2004, Gray and Basmajian 1968). Some studies have used the medial longitudinal arch height as an indicator of foot pronation (Franettovich et al 2008), however this might not be the best method of classifying foot type as it was observed in the present study that pronation was not always related to a low medial arch, except in one participant who had the greatest ND (2.03 mm). Neely (1998) supported this observation in his review stating that a pes planus foot (low medial longitudinal arch) is always accompanied by foot pronation but that the opposite is not true.
A gait analysis treadmill was used to standardise walking speed and cadence during the trial and to overcome the problem of insufficient EMG lead length. However, walking on a treadmill has been shown to have minor effects on gait parameters like the stance period and cadence (Warabi et al 2005) and subtalar joint biomechanics during stance phase of gait (Sajko and Pierrynowski 2005). Although, quantitative and qualitative characteristics and overall pattern of gait on a treadmill have been found to be very similar to walking over ground (Riley et al 2007) the use of a treadmill could have resulted in minor alterations to gait in a way which could have changed TA muscular activity (Lee and Hindler 2008) and consequently affected the results of this study.
Average speeds for males and females between the ages of 20-59 years were used in the present study. An attempt to calculate the speed of each participant was done in a pilot study but the over ground speed of subjects did not correspond to the same velocity when set on the treadmill. The treadmill’s belt speed was much slower, therefore preset speeds were chosen. This might be considered a limitation as the predetermined velocity might have been slightly slower or faster than the participant’s preferred walking velocity. This could have had an effect on TA muscle EMG output, since an earlier study by Nilsson et al (1985) identified that TA EMG activity was shown to increase with increasing velocity (with each 0.2 m.s-1 increment between 0.4 and 3.0 m.s-1). Furthermore, the timing of muscle EMG activity was speed dependant (Stoquart et al. 2008). Although conversely, TA muscle EMG activity was found to be stable with the speed changes except at very slow velocities (down to 0.28 m.s-1) (den Otter et al. 2004) which is substantially slower than the treadmill speeds used in this study.
The results of this pilot study are not sufficient to support the hypothesis that foot type is a mediator of TA activity. Therefore there is no evidence to support the view that foot type will lead to lower limb injury as a consequence of altered muscle activity as was found by previous studies (Kaufman et al 1999, Bennett et al 2001). However, differences in the methodology of foot classification between these studies and the current study may have lead to differences in results between the studies.
Future research should consider including larger number of participants with more extreme ND values of pronated and supinated foot types, which could reveal the effect of foot type on TA muscle EMG activity if there was any. This is recommended since no power calculation was performed in this study to provide an accurate estimation of required sample size to detect differences in muscle EMG activity. Future studies investigating the effect of foot type on muscle EMG activity during gait could consider utilizing over ground walkways using the participants’ preferred walking speed which would allow observation of TA EMG activity where subjects are able to employ their usual foot and lower limb mechanics. It would also be recommended to utilize a dynamic foot posture assessment tool using video cameras and markers in addition to a static foot posture measurement. Identifying the static posture in weight bearing as was done in this study through the NDT might give an idea about the foot posture during gait, since it might resemble the stance phase but it may not be identical to it, so more accurate dynamic assessment of foot position during walking might be more appropriate.
This pilot study investigated the effect of foot type on TA muscle activity as measured by surface EMG and found no significant differences in the activity of TA during any of the stance phases of gait between asymptomatic subjects with pronated, normal or supinated feet. Therefore there is no evidence to support the view that foot type will lead to lower limb injury as a consequence of altered TA muscle activity. In conclusion, foot type is unlikely to be an etiological factor of TA muscle injuries.
Beck BR (1998) Tibial stress injuries: An aetiological review for the purposes of guiding management. Sports medicine 26(4);265-279
Bennett JE, Reinking MF, Pluemer B, Pentel A, Seaton M and Killian C (2001) Factors contributing to the development of medial tibial stress syndrome in high school runners. The journal of orthopaedic and sports and physical therapy 31(9);504-510
Bogey R, Cerny K and Mohammed O (2003) Repeatability of wire and surface electrodes in gait. American journal of physical medicine and rehabilitation 82(5);338-344
Brody DM (1982) Techniques in the evaluation and treatment of the injured runner. The orthopedic clinics of north America 13(3);541-558
Cote KP, Brunet ME, Gasneder BM, and Shultz SJ (2005) Effects of pronated and supinated foot postures on static and dynamic postural stability. Journal of athletic training 40(1);41-46
Cornwall MW and McPoil TG (1994) The influence of tibialis anterior muscle activity on rearfoot motion during walking. Foot & ankle international 15(2);75-79
Cowan DN, Robinson JR, Jones BH, Polly DW and Berry BH (1994) Consistency of visual assessments of arch height among clinicians. Foot & ankle international 15(4);213-217
De Lacerda FG (1980) A study of anatomical factors involved in shinsplints. The journal of orthopaedic and sports and physical therapy 2(2);5-59
den Otter AR, Geurts AC, Mulder T and Duysens J (2004) Speed related changes in muscle activity from normal to very slow walking speeds. Gait & posture 19(3);270-278
Donatelli R (1985) Normal biomechanics of the foot and ankle. The journal of orthopaedic and sports and physical therapy 7(3);91-95
Donatelli R (1987) Abnormal biomechanics of the foot and ankle. The journal of orthopaedic and sports and physical therapy 9(1);11-16
Field A (2005) Discovering statistics using SPSS, second ed. London, Sage
Franettovich M, Chapman A and Vicenzino B (2008) Tape that increases medial longitudinal arch height also reduces leg muscle activity: a preliminary study. Medicine and science in sports and exercise 40(4);593-600
Gray EG and Basmajian JV (1968) Electromyography and cinematography of leg and foot (“normal and flat”) during walking. The anatomical record 161(1);1-16
Hunt AE and Smith RM (2004) Mechanics and control of the flat versus normal foot during the stance phase of walking. Clinical biomechanics 19(4);391-397
Hunt AE, Fahey AJ and Smith RM (2000) Static measures of calcaneal deviation and arch angle as predictors of rearfoot motion during walking. The Australian journal of physiotherapy 46(1);9-16
Kaufman KR, Brodine SK, Shaffer RA, Johnson CW and Cullison TR (1999) The effect of foot structure and range of motion on musculoskeletal overuse injuries. The American journal of sports medicine 27(5);585-593
Kistler Instrument Corporation (2006) Gaitway Treadmill Type 9810A, Instrumented Treadmill for Gait Analysis. [Online] Available from: http://www.kistler.com/mediaaccess/us/000-174a-01.06.pdf?sid=c0a86c27232989d5dda70ba14501afc5e206b9ac7c31&location=http://www.kistler.com/do.searchResult.us.en-us?content=searchresult¶m=gaitway+software. [Accessed 18 August 2008]
Kram R, Griffin TM, Donelan JM and Chang YH (1998) Force treadmill for measuring vertical and horizontal ground reaction forces. Journal of applied physiology 85(2) 764-769
Loudon JK, Jenkins W and Loudon KL (1996) The relationship between static posture and ACL injury in female athletes. The journal of orthopaedic and sports and physical therapy 24(2);91-97
Lee SJ and Hidler J (2008) Biomechanics of overground vs. treadmill walking in healthy individuals. Journal of applied physiology 104(3);747-755
Menz HB, Morris ME and Lord SR (2005) Foot and ankle characteristics associated with impaired balance and functional ability in older people. The journals of gerontology. Series A, Biological sciences and medical sciences 60(12);1546-1552
Neely FG (1998) Biomechanical risk factors for exercise-related lower limb injuries. Sports medicine 26(6);395-413
Nilsson J, Thorstensson A and Halbertsma J (1985) Changes in leg movements and muscle activity with speed of locomotion and mode of progression in humans. Acta Physiologica Scandinavica 123;457-475
O’Connor KM and Hawill J (2004) The role of extrinsic foot muscles during running. Clinical biomechanics 19;71-77
Razeghi M and Batt ME (2002) Foot type classification: A critical review of current methods. Gait & posture 15;282-291
Riley PO, Paolini G, Della Croce U, Paylo KW and Kerrigan DC (2007) A kinematic and kinetic comparison of overground and treadmill walking in healthy subjects. Gait & posture 26;17-24
Sajko SS and Pierrynowski MR (2005) Influence of treadmill design on rearfoot pronation during gait at different speeds. Acta Physiologica Scandinavica 95;475-80
Sell KE, Verity TM, Worrell TW, Pease BJ and Wigglesworth J (1994) Two measurement techniques for assessing subtalar joint position: A reliability study. The journal of orthopaedic and sports and physical therapy 19;162-167
Snook AG (2001) The relationship between excessive pronation as measured by navicular drop and isokinetic strength of the ankle musculature. Foot & ankle international 22;234-240
Stoquart G, Detrembleur C, Lejeune T (2008) Effect of speed on kinematic, kinetic, electromyographic and energetic reference values during treadmill walking. Clinical neurophysiology 38, 105-116
Trimple MH, Bishop MD, Buckley BD, Fields LC and Rozea GD (2002) The relationship between clinical measurements of lower extremity and tibial translation. Clinical biomechanics 17;286-290
Warabi T, Kato M, Kiriyama K, Yoshida T and Kobayashi N (2005) Treadmill walking and overground walking of human subjects compared by recording sole-floor reaction force. Neuroscience research 53;343-348
Waters RL, Lunsford BR, Perry J and Byrd R (1988) Energy-speed relationship of walking: standard tables. Journal of orthopaedic research 6;215-222
Whittle MW (2002) Gait analysis: an introduction, third ed. Oxford, Butterworth-Heinemann
Williams DS and McClay IS (2000) Measurements used to characterize the foot and the medial longitudinal arch: Reliability and validity. Physical therapy 80;864-871