Gait & Posture 39 (2014) 1080–1085

Contents lists available at ScienceDirect

Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost

Electromyographic patterns of tibialis posterior and related muscles when walking at different speeds George S. Murley a,b,*, Hylton B. Menz a,b, Karl B. Landorf a,b a b

Department of Podiatry, Faculty of Health Sciences, La Trobe University, Bundoora 3086, Australia Lower Extremity and Gait Studies Program, Faculty of Health Sciences, La Trobe University, Bundoora 3086, Australia

A R T I C L E I N F O

A B S T R A C T

Article history: Received 23 September 2013 Received in revised form 4 December 2013 Accepted 22 January 2014

The effect of walking speed on superficial lower limb muscles, such as tibialis anterior and triceps surae, is well established. However, there are no published data available for tibialis posterior – a muscle that plays an important role in controlling foot motion. The purpose of this study was to characterise the electromyographic timing and amplitude of selected lower limb muscles across five walking speeds. Thirty young adults were instructed to walk barefoot while electromyographic activity was recorded from tibialis posterior and peroneus longus via intramuscular electrodes, and medial gastrocnemius and tibialis anterior via surface electrodes. At faster walking speeds, peak electromyographic amplitude increased systematically during the contact and midstance/propulsion phases. Changes in the time of peak amplitude were also observed for tibialis posterior, tibialis anterior and peroneus longus activity; however, these were muscle and phase specific. During contact phase, peak electromyographic amplitude for tibialis posterior and peroneus longus was similar across very slow to slow walking speeds. During midstance/propulsion phase, peak electromyographic amplitude for tibialis posterior and medial gastrocnemius was similar across very slow to slow walking speeds. These findings may reflect a relatively higher than expected demand for peroneus longus and tibialis posterior to assist with mediolateral foot stability at very slow speeds. Similarly, peak amplitude of medial gastrocnemius was also relatively unchanged at the very slow speed, presumably to compensate for the reduced forward momentum. The data presented in this study may serve as a reference for comparing similarly matched participants with foot deformity and/or pathological gait. ß 2014 Elsevier B.V. All rights reserved.

Keywords: Gait Speed Posture Muscle activity

1. Introduction The speed-dependent nature of several gait parameters is well established in the literature [1]. For example, electromyography (EMG) amplitude systematically changes (i.e. linear/curvilinear) for lower limb muscles with walking speeds ranging from very slow to very fast [1–6]. For the majority of lower limb muscles, EMG amplitude generally increases with faster than normal speeds and decreases with slower than normal speeds. These speed-dependent changes may be important as slower freely chosen walking speeds are associated with several neurological and rheumatological diseases [7,8]. Therefore, when comparing gait of individuals with pathology to that of healthy individuals, researchers must account for differences in muscle activity related to disease progression or pain versus differences

* Corresponding author at: Department of Podiatry and Lower Extremity and Gait Studies Program, Faculty of Health Sciences, La Trobe University, Bundoora, VIC 3086, Australia. Tel.: +61 3 9479 5834; fax: +61 3 9479 5768. E-mail address: [email protected] (G.S. Murley). 0966-6362/$ – see front matter ß 2014 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.gaitpost.2014.01.018

related to walking speed (i.e. changes from disease compared to changes from walking speed). To assist with this process, it is fundamental that we understand how individual muscles behave at different walking speeds in healthy individuals. While several studies have reported EMG profiles for superficial muscles, such as tibialis anterior and triceps surae [1–3,5,9–11], to our knowledge there are no published data available for tibialis posterior – a muscle that plays an important role in controlling foot motion and one that is affected by the diseases stated above. In addition to characterising the activity of individual muscles at different speeds to obtain normative data, it is also of interest to investigate whether the function of individual muscles change at different speeds. For example, at very slow speeds, it is thought that relatively greater muscle activity is required to maintain medio-lateral postural stability. This was demonstrated by den Otter and colleagues [5] who found that at very slow speeds, compared to moderately slow speeds, there is increased normalised EMG amplitude of peroneus longus and rectus femoris in stance phase and swing phase, respectively. Despite the abundance of gait research related to the effects of walking speed on lower limb biomechanics, there are several

G.S. Murley et al. / Gait & Posture 39 (2014) 1080–1085

methodological limitations that need to be addressed. One limitation relates to the use of treadmill rather than overground walking in previous studies related to walking speed [2,3,5,6,9,11,12]. As treadmill walking is known to alter gait parameters [13], it is preferable (where practical), to measure gait parameters during overground walking. Other issues with previous research include the use of small sample sizes, and an imbalance of male and female participants in studies. These issues all reduce the ability to generalise the results to the wider population. Therefore, the aim of this study was to investigate the effect of a wide range of self-selected walking speeds on tibialis posterior, peroneus longus, tibialis anterior and medial gastrocnemius during overground walking. 2. Methods 2.1. Participants Thirty young adults were recruited into the study after providing informed consent (see Table 1 for participant characteristics). Participants had no recent musculoskeletal symptoms and denied having a neuromuscular disease. Only those with normalarched feet were included in this study, as foot posture is known to influence lower limb muscle activity [14]. A foot screening protocol was used to assess foot posture that included both clinical and radiographic measures [15]. Ethical approval was obtained from

[(Fig._1)TD$IG]

120

1081

Table 1 Participant characteristics (N = 30). Gender ratio (female/male) Age in years, mean [SD] Height in cm, mean [SD] Mass in kg, mean [SD]

15/15 23.6 [5.9] 169.7 [9.7] 69.9 [13.6]

the La Trobe University Human Ethics Committee (Ethics ID: FHEC06/205). 2.2. Instrumentation Tibialis posterior (TP) and peroneus longus (PL) were recorded using bipolar fine-wire intramuscular electrodes. The electrodes were fabricated from 75 mm Teflon1 coated stainless steel wire (AM Systems, Washington, USA) with 1 mm of insulation stripped to form the recording surface of the two wires. The electrode wires were inserted into a single-use 23-gauge hypodermic needle with the exposed electrode tips bent 3 mm and 5 mm (to prevent the contact areas from touching during recording) and were sterilised prior to use. The processes of fine-wire electrode construction and positioning of wires in vivo were undertaken in accordance with previous work [16]. Tibialis anterior (TA) and medial gastrocnemius (MG) EMG signals were recorded with the use of DE-3.1 surface electrodes (Delsys Inc., Boston, USA). The electrodes featured a double differential 3-bar type configuration with a 99.9% silver contact 120

Tibialis posterior

Peroneus longus

110 100

100

Fast

80

Normal

70

Slow Very Slow

60 50 40 30

Very Fast

EMG amplitude (% MVIC)

EMG amplitude (% MVIC)

Very Fast 90

Fast 80

Normal Slow Very Slow

60

40

20

20 10 0

0 0

10

20

30

40

50

60

70

80

90

0

100

10

20

30

% Gait cycle 120

50

60

70

80

90

100

% Gait cycle 120

Tibialis anterior

110

Medial gastrocnemius

110

100

100

90

Fast

80

Normal

70

Slow Very Slow

60 50 40 30

EMG amplitude (% MVIC)

Very Fast

EMG amplitude (% MVIC)

40

Fast Normal

80

Slow 70

Very Slow

60 50 40

20

30

10

20

0

Very Fast

90

10 0

10

20

30

40

50

60

% Gait cycle

70

80

90

100

0

10

20

30

40

50

60

70

80

90

100

% Gait cycle

Fig. 1. EMG ensemble averages derived from a single gait cycle for each participant. The curves differ slightly to the actual results as these curves are derived from a single gait cycle for each participant to illustrate the main findings. 0% – heel contact; 100% – ipsilateral heel contact. Numerical data for Fig. 1 is available in Supplementary file 3.

1082

G.S. Murley et al. / Gait & Posture 39 (2014) 1080–1085

material and an inter-electrode distance of 10 mm. The application of surface electrodes followed the recommendations of SENIAM – Surface Electromyography for the Non-Invasive Assessment of Muscles [17]. The temporal events of the walking cycle (heel contact, toe contact, heel off and toe off) were measured using circular forcesensitive resistors (footswitches), with a diameter of 13 mm (Model: 402, Interlink Electronics, CA, USA).

2.3.1. EMG normalisation technique At the completion of the gait trials, three maximum isometric voluntary contractions (MVICs) for each muscle were undertaken to normalise the amplitude data as described in earlier work [16]. These comprised a gradual and continuous 2-s build-up followed by a maximum 2-s effort. Three consecutive maximum efforts were separated by a 1-min recovery period. A 600 ms window in the middle of the 2-s recording period was used to calculate average root mean square (RMS) from three trials.

2.3. Data collection protocol During testing, participants were instructed to walk at the following five randomly allocated self-selected walking speeds whilst barefoot, using the descriptors reported by Latt et al. [18]: (i) much slower than usual, as though they were strolling in the park – referred to as ‘very slow’ (ii) slightly slower than usual – referred to as ‘slow’ (iii) their usual comfortable walking speed – referred to as ‘normal’ (iv) slightly faster than usual (as though they were in a hurry) – referred to as ‘fast’ (v) as fast as possible without running (as though they were late for an appointment) – referred to as ‘very fast’ In our study, the participants’ self-selected speeds were established during a warm-up period from two trials along a nine metre walkway. Six trials were recorded during testing to ensure consistency of average speed. Any trial exceeding 5% of the average warm-up speed was excluded, with the trial being repeated. [(Fig._2)TD$IG]

2.3.2. EMG processing Raw EMG signals were passed through a differential amplifier (Delsys Inc., Boston, USA; input impedance = 1015 X//0.2 pF, CMRR = 92 dB @ 60 Hz) at a gain of 1000. Band pass filtering (built into the amplifier) of 20–2000 Hz was applied for intramuscular electrodes and 20–450 Hz for surface electrodes and sampled at 2000 Hz. Two consecutive strides (i.e. comprising three consecutive heel contacts from the ipsilateral limb) were analysed for each trial and averaged from the last four of six trials for each speed (i.e. four average gait cycles derived from eight ipsilateral steps). Two EMG parameters were analysed for each muscle: (i) time of peak amplitude and (ii) normalised peak amplitude. These parameters have been utilised in previous single-session investigations [14,19,20]. Recorded EMG data were full wave rectified and low pass filtered at a cut off frequency of 6 Hz through a 4th order Butterworth filter with phase lag to best represent muscle tension through the gait cycle [21,22]. The following phases of the gait cycle were assessed based on when each muscle is most active in normal-

Fig. 2. Tibialis posterior – scatter plots with regression lines for each speed and EMG parameter. Solid vertical and horizontal error bars represent 95% confidence intervals. Broken horizontal lines indicate significant differences between walking speeds (p < 0.05).

G.S. Murley et al. / Gait & Posture 39 (2014) 1080–1085

1083

arched feet [16]: contact and combined midstance/propulsion phase for TP and PL; contact phase for TA; and combined midstance/ propulsion phase for MG. Contact phase was defined as the period between heel contact and toe contact; while combined midstance/ propulsion was defined as the period between toe contact and toe off.

F-ratio and degrees of freedom were taken from the Greenhouse– Geisser epsilon (a = 0.05). To account for multiple comparisons, statistically significant univariate F-statistics were evaluated with Bonferroni-adjusted pairwise comparisons, which were considered significant when P < 0.05.

2.4. Statistical analysis

3. Results 3.1. Walking speed

To ensure normality of the data, the distribution of all EMG variables were initially assessed with skewness and kurtosis values and by evaluating distribution curves from histograms. To test for differences between conditions, a series of one-way repeated measures ANOVA tests were conducted separately for each muscle, phase of gait, and EMG parameter. The series of within-subject factors for each muscle analysis are summarised below: (i) TP – two phases of gait  two EMG parameters (four one-way ANOVA tests). (ii) PL – two phases of gait  two EMG parameters (four one-way ANOVA tests). (iii) TA – one phase of gait  two EMG parameters (two one-way ANOVA tests). (iv) MG – one phase of gait  two EMG parameters (two one-way ANOVA tests). Where data violated the assumption for sphericity as determined by significant results (P < 0.05) for the Mauchley’s test, the

[(Fig._3)TD$IG]

The participants’ self-selected walking speed increased linearly from the very slow to very fast walking conditions [mean (SD) very slow – 0.67 (0.11) m/s, slow – 0.85 (0.18) m/s, normal – 1.10 (0.10) m/s, fast – 1.29 (0.14) m/s, very fast – 1.63 (0.18) m/s)]. These speeds only varied slightly to those reported previously (Latt et al. [18] who used the same instructions for walking speed) by 0.17 ms and 0.47 ms for the very slow and very fast walking speeds, respectively. 3.2. EMG muscle activity Mean data for each walking speed for time of peak and peak EMG amplitude are presented in Supplementary files 1 and 2. Fig. 1 presents EMG ensemble averaged tracings and Figs. 2–4 present scatter plots for all muscles and conditions with statistically significant findings indicated. 3.2.1. Contact phase EMG activity Significant within-participant effects were detected for the three muscles assessed during contact phase (TP, PL and TA). Multiple pair-wise comparisons between conditions revealed several significant findings for TP, PL and TA time of peak and peak EMG amplitude (Figs. 2–4). A significant and systematic increase in peak EMG amplitude was detected with faster speeds for TP, PL and TA. No significant differences were detected for TP and

Fig. 3. Peroneus longus – scatter plots with regression lines for each speed and EMG parameter. Solid vertical and horizontal error bars represent 95% confidence intervals. Broken horizontal lines indicate significant differences between walking speeds (p < 0.05).

[(Fig._4)TD$IG]

1084

G.S. Murley et al. / Gait & Posture 39 (2014) 1080–1085

Fig. 4. Tibialis anterior and medial gastrocnemius – scatter plots with regression lines for each speed and EMG parameter. Solid vertical and horizontal error bars represent 95% confidence intervals. Broken horizontal lines indicate significant differences between walking speeds (p < 0.05). PL comparing very slow and slow speeds. With faster walking speeds, a significant and systematic delay was detected for TP and TA time of peak amplitude. A delay in time of peak amplitude was detected for PL with the very fast walking speed compared to the slow speed.

3.2.2. Midstance/propulsion phase EMG activity Significant within-participant effects were detected for the three muscles assessed during midstance/propulsion phase (TP, PL and MG). Multiple pair-wise comparisons between conditions revealed several significant findings for TP, PL and MG peak EMG amplitude. For PL, changes were only detected for time of peak amplitude (Figs. 2–4). A significant and systematic increase in peak EMG amplitude was detected for TP, PL and MG with faster speeds. No significant differences were detected for TP and MG comparing very slow and slow speeds. With slower walking speeds, a significant and systematic delay was detected for PL time of peak amplitude. No significant differences were detected for TP or MG time of peak amplitude comparing the five walking speeds.

4. Discussion The aim of this study was to characterise the EMG timing and amplitude of selected lower limb muscles across five walking speeds. As with previous research, we found muscle activity was strongly affected by walking speed with significant and systematic gains in EMG peak amplitude as walking speed increased [1–5]. Significant temporal changes in peak amplitude were also detected for TP, PL and TA, however these were muscle and phase-dependant (i.e. the direction of change was unique to either contact or midstance phase for each muscle). The kinematic/kinetic mechanism causing latency of TP, PL and TA with increasing speeds

during contact phase is difficult to explain, as this period of gait is very short (less than 20% of gait) and little is known about speed dependency of movement patterns and forces acting around the medial arch and midfoot regions during this period. Earlier activation of PL with faster speeds during midstance might be associated with decreased and earlier maximal dorsiflexion of the forefoot during this phase [23]. Despite the shifts in time of peak amplitude with different walking speeds, none of the muscles displayed a complete phasic change or loss of activity with any particular speed. Whilst several aspects of the data presented in this study confirm previous reports, i.e. biphasic and speed dependent EMG for PL [5,9,24], some new and important findings were apparent. This is the first study to investigate speed-related changes in TP EMG muscle activity. Fig. 1 shows that TP maintained the typical two-burst profile during contact and midstance/propulsion phases, irrespective of speed, and displayed significant gains in peak EMG amplitude with increasing walking speed (Fig. 2). Interestingly, TP peak EMG amplitude was similar at slow to very slow speeds during both contact and midstance/propulsion phase. This finding may reflect a relatively higher than expected demand for TP to assist with medio-lateral foot stability at very slow speeds, presumably when the medial arch is more mobile during extended periods of single support. Like previous reports [3,10], MG peak EMG amplitude was similar at slow to very slow speeds during midstance/propulsion phase. Therefore, it is evident that even when walking speed is decreased to a very slow speed with very little forward momentum, the triceps surae musculature is vital to the production of forward motion.

G.S. Murley et al. / Gait & Posture 39 (2014) 1080–1085

The finding that peak EMG amplitude was similar at very slow and slow walking speeds for TP is mirrored by PL EMG activity during contact phase. This may be important since TP and PL are considered to have both antagonistic (i.e. medio-lateral foot stability) and synergistic (i.e. ankle joint plantarflexion) relationships. The dependency for TP and PL to provide additional stability at very slow speed should be explored further with running gait and in conjunction with other EMG parameters associated with stability, such as averaged amplitude. Another notable finding with respect to TP and PL were the relative changes in time of peak amplitude during midstance/ propulsion phase. Fig. 2 indicates that time of peak amplitude for TP is relatively unchanged by different speeds, whereas Fig. 3 indicates a systematic shift in the direction towards earlier time of peak amplitude for PL with increasing speed. The finding that PL time of peak amplitude is affected by walking (and not TP) may indicate PL has a larger role in maintaining foot stability during midstancepropulsion. Further research is needed to investigate whether this pattern is altered in those with recurrent ankle instability, given that delayed PL latency in response to standing perturbation has previously been reported in these individuals [25]. The findings from this study are of importance for two reasons. Firstly, the data presented are from predominantly younger adults with normal-arched foot posture – this may serve as a reference for comparing similarly matched individuals with foot deformity and or pathological gait. Secondly, the new data for TP serves to inform researchers investigating individuals with pathological gait about the potential confounding effects of walking speed as individuals with pain and/or disability may walk slower to relieve pain, or because of a functional impairment or poor balance. 4.1. Limitations The findings from this study need to be viewed in light of two limitations. Firstly, we did not simultaneously collect kinematic or kinetic data from participants. Therefore, we can only speculate about the specific functions of the muscles tested during different walking speeds. Secondly, as the MVICs were conducted manually (i.e. without using an isokinetic dynamometer), some random variability may be attributed to the tester resisting the MVICs (e.g. medial gastrocnemius is a powerful muscle that is difficult to resist with manual testing). 5. Conclusion In summary, our results demonstrate that muscle activity was strongly affected by walking speed. Peak EMG amplitude for TP, PL and MG were similar across very slow to slow walking speeds, presumably to assist with medio-lateral foot stability and to compensate for reduced forward momentum at very slow speeds. These results may be useful for comparing individuals with pathological gait. Conflicts of interest statement The authors have no conflicts of interest to declare. Acknowledgements This study was funded by the Australian Podiatry Education and Research Fund (APERF). HBM is currently a National Health and Medical Research Council Senior Research Fellow (ID: 1020925).

1085

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.gaitpost.2014. 01.018. References [1] Schwartz MH, Rozumalski A, Trost JP. The effect of walking speed on the gait of typically developing children. J Biomech 2008;41:1639–50. [2] Anders C, Wagner H, Puta C, Grassme R, Petrovitch A, Scholle HC. Trunk muscle activation patterns during walking at different speeds. J Electromyogr Kinesiol 2007;17:245–52. [3] Warren GL, Maher RM, Higbie EJ. Temporal patterns of plantar pressures and lower-leg muscle activity during walking: effect of speed. Gait Posture 2004;19:91–100. [4] Chiu MC, Wang MJ. The effect of gait speed and gender on perceived exertion, muscle activity, joint motion of lower extremity, ground reaction force and heart rate during normal walking. Gait Posture 2007;25:385–92. [5] den Otter AR, Geurts AC, Mulder T, Duysens J. Speed related changes in muscle activity from normal to very slow walking speeds. Gait Posture 2004;19: 270–8. [6] van Hedel HJ, Tomatis L, Muller R. Modulation of leg muscle activity and gait kinematics by walking speed and bodyweight unloading. Gait Posture 2006;24:35–45. [7] Patterson KK, Gage WH, Brooks D, Black SE, McIlroy WE. Changes in gait symmetry and velocity after stroke: a cross-sectional study from weeks to years after stroke. Neurorehabil Neural Repair 2010;24:783–90. [8] Turner DE, Helliwell PS, Siegel KL, Woodburn J. Biomechanics of the foot in rheumatoid arthritis: identifying abnormal function and the factors associated with localised disease ‘impact’. Clin Biomech 2008;23:93–100. [9] Ivanenko YP, Poppele RE, Lacquaniti F. Five basic muscle activation patterns account for muscle activity during human locomotion. J Physiol 2004;556: 267–82. [10] Nymark JR, Balmer SJ, Melis EH, Lemaire ED, Millar S. Electromyographic and kinematic nondisabled gait differences at extremely slow overground and treadmill walking speeds. J Rehabil Res Dev 2005;42:523–34. [11] Stoquart G, Detrembleur C, Lejeune T. Effect of speed on kinematic, kinetic, electromyographic and energetic reference values during treadmill walking. Neurophysiol Clin 2008;38:105–16. [12] Clancy EA, Cairns KD, Riley PO, Meister M, Kerrigan DC. Effects of treadmill walking speed on lateral gastrocnemius muscle firing. Am J Phys Med Rehabil 2004;83:507–14. [13] Lee SJ, Hidler J. Biomechanics of overground vs. treadmill walking in healthy individuals. J Appl Physiol 2008;104:747–55. [14] Murley GS, Menz HB, Landorf KB. Foot posture influences the electromyographic activity of selected lower limb muscles during gait. J Foot Ankle Res 2009;2:35. [15] Murley GS, Menz HB, Landorf KB. A protocol for classifying normal- and flatarched foot posture for research studies using clinical and radiographic measurements. J Foot Ankle Res 2009;2:22. [16] Murley GS, Buldt AK, Trump PJ, Wickham JB. Tibialis posterior EMG activity during barefoot walking in people with neutral foot posture. J Electromyogr Kinesiol 2009;19:e69–77. [17] Hermens HJ, Freriks B, Merletti R, Stegeman D, Blok J, Rau G, et al. SENIAM 8 – European recommendations for surface electromyography. 2nd ed. Enschede: Roessingh Research and Development; 1999, 121 pp.. [18] Latt MD, Menz HB, Fung VS, Lord SR. Walking speed, cadence and step length are selected to optimize the stability of head and pelvis accelerations. Exp Brain Res 2008;184:201–9. [19] Murley GS, Landorf KB, Menz HB. Do foot orthoses change lower limb muscle activity in flat-arched feet towards a pattern observed in normal-arched feet? Clin Biomech 2010;25:728–36. [20] Murley GS, Menz HB, Landorf KB, Bird AR. Reliability of lower limb electromyography during overground walking: a comparison of maximal- and submaximal normalisation techniques. J Biomech 2010;43:749–56. [21] Semciw AI, Pizzari T, Murley GS, Green RA. Gluteus medius: an intramuscular EMG investigation of anterior, middle and posterior segments during gait. J Electromyogr Kinesiol 2013;23:858–64. [22] Willson JD, Petrowitz I, Butler RJ, Kernozek TW. Male and female gluteal muscle activity and lower extremity kinematics during running. Clin Biomech 2012;27:1052–7. [23] Tulchin K, Orendurff M, Adolfsen S, Karol L. The effects of walking speed on multisegment foot kinematics in adults. J Appl Biomech 2009;25: 377–86. [24] Hof AL, Elzinga H, Grimmius W, Halbertsma JP. Speed dependence of averaged EMG profiles in walking. Gait Posture 2002;16:78–86. [25] Hopkins JT, Brown TN, Christensen L, Palmieri-Smith RM. Deficits in peroneal latency and electromechanical delay in patients with functional ankle instability. J Orthop Res 2009;27:1541–6.

Electromyographic patterns of tibialis posterior and related muscles when walking at different speeds.

The effect of walking speed on superficial lower limb muscles, such as tibialis anterior and triceps surae, is well established. However, there are no...
1MB Sizes 1 Downloads 3 Views