Introduction

Gait initiation is a temporary movement between upright posture and steady-state gait. The activation of several postural muscles has been identified to precede changes observed in vertical reaction force (Winter, 1995). Although previous research has focused on the lower limb, few studies have examined recruitment patterns of the thigh and trunk musculature. This study was conducted to determine the phasic patterns of muscles of the lower limbs and trunk for the duration from quiet stance to trail-limb toe-off.

Introduction

Gait initiation is a temporary movement between upright posture and steady-state gait. The activation of several postural muscles has been identified to precede changes observed in vertical reaction force (Winter, 1995). Although previous research has focused on the lower limb, few studies have examined recruitment patterns of the thigh and trunk musculature. This study was conducted to determine the phasic patterns of muscles of the lower limbs and trunk for the duration from quiet stance to trail-limb toe-off.

Methods

Eleven healthy participants initiated gait with their right legs. Two force platforms (Kistler) were used to measure vertical ground reaction forces (GRF), at 1040 Hz, from quiet stance to toe-off of the trail limb. In addition, electromyographic data (Octopus, Bortec) were collected at 1040 Hz beginning at quiet stance to the end of the third step. EMG electrodes were placed bilaterally over the erector spinae (ES), the tensor fasciae latae (TFL), the adductor magnus (ADD) and the tibialis anterior (TA) muscles.

Participants stood with one foot on each plate, distributing their weight equally. Each participant walked briskly, after the researcher gave a “go” command. Ten trials were collected for each subject. Force platform data were filtered with a zero-lag, second-order, critically damped, low-pass filter with a cut-off frequency of 20 Hz. To remove low frequency motion artefacts, the raw electromyographic data were high-pass filtered with a cut-off frequency of 8 Hz (Robertson & Dowling, 2003). Electromyographic data were full-wave rectified and filtered by a second-order, critically damped, low-pass filter with a cut-off frequency of 5 Hz, producing a linear envelope (Robertson & Dowling, 2003).

An amplitude threshold criterion determined the start and end of muscle activity. The threshold was based on three times the standard deviation of the EMG during quiet stance for each muscle, estimated from the least variable 100 ms period of each EMG. Timings of all eight muscle onsets and offsets were recorded from time-normalized, ensemble-averaged data for each subject for the period beginning 1.5 s before trail-limb toe- off.

Methods

Eleven healthy participants initiated gait with their right legs. Two force platforms (Kistler) were used to measure vertical ground reaction forces (GRF), at 1040 Hz, from quiet stance to toe-off of the trail limb. In addition, electromyographic data (Octopus, Bortec) were collected at 1040 Hz beginning at quiet stance to the end of the third step. EMG electrodes were placed bilaterally over the erector spinae (ES), the tensor fasciae latae (TFL), the adductor magnus (ADD) and the tibialis anterior (TA) muscles.

Participants stood with one foot on each plate, distributing their weight equally. Each participant walked briskly, after the researcher gave a “go” command. Ten trials were collected for each subject. Force platform data were filtered with a zero-lag, second-order, critically damped, low-pass filter with a cut-off frequency of 20 Hz. To remove low frequency motion artefacts, the raw electromyographic data were high-pass filtered with a cut-off frequency of 8 Hz (Robertson & Dowling, 2003). Electromyographic data were full-wave rectified and filtered by a second-order, critically damped, low-pass filter with a cut-off frequency of 5 Hz, producing a linear envelope (Robertson & Dowling, 2003).

An amplitude threshold criterion determined the start and end of muscle activity. The threshold was based on three times the standard deviation of the EMG during quiet stance for each muscle, estimated from the least variable 100 ms period of each EMG. Timings of all eight muscle onsets and offsets were recorded from time-normalized, ensemble-averaged data for each subject for the period beginning 1.5 s before trail-limb toe- off.

Muscle Activation Patterns During Gait InitiationMuscle Activation Patterns During Gait InitiationNatasha Kyle, MSc and D. Gordon E. Robertson, PhD, FCSBNatasha Kyle, MSc and D. Gordon E. Robertson, PhD, FCSB

School of Human Kinetics, University of Ottawa, Ontario, CanadaSchool of Human Kinetics, University of Ottawa, Ontario, Canada

Muscle Activation Patterns During Gait InitiationMuscle Activation Patterns During Gait InitiationNatasha Kyle, MSc and D. Gordon E. Robertson, PhD, FCSBNatasha Kyle, MSc and D. Gordon E. Robertson, PhD, FCSB

School of Human Kinetics, University of Ottawa, Ontario, CanadaSchool of Human Kinetics, University of Ottawa, Ontario, Canada

Results

Table 1 holds the muscle onset times for the nine subjects. The earliest muscle activated was consistently the lead-limb tibialis anterior, followed by the lead-limb tensor fasciae latae (figure 1). The trail-limb tibialis anterior, trail-limb tensor fasciae latae and the trail-limb adductor magnus were next to become active, respectively. The order of muscle activity during the middle of the gait initiation process varied. Specifically, there were notable inconsistencies between subjects for the order of the fifth and sixth muscle activations. The last two muscles to activate were consistently the erector spinae of the trail-limb side followed by the erector spinae of the lead-limb.

Results

Table 1 holds the muscle onset times for the nine subjects. The earliest muscle activated was consistently the lead-limb tibialis anterior, followed by the lead-limb tensor fasciae latae (figure 1). The trail-limb tibialis anterior, trail-limb tensor fasciae latae and the trail-limb adductor magnus were next to become active, respectively. The order of muscle activity during the middle of the gait initiation process varied. Specifically, there were notable inconsistencies between subjects for the order of the fifth and sixth muscle activations. The last two muscles to activate were consistently the erector spinae of the trail-limb side followed by the erector spinae of the lead-limb.

Biomechanics Laboratory

ReferencesRobertson DGE & Dowling JJ (2003) J Electromyo Kines, 13: 569-573.Robertson DGE, Smith, O’Dwyer (2005) Proceed ISB XX, p.102.Winter DA (1995) A.B.C. of Balance during Standing and

Walking. Waterloo: Waterloo Biomechanics.

ReferencesRobertson DGE & Dowling JJ (2003) J Electromyo Kines, 13: 569-573.Robertson DGE, Smith, O’Dwyer (2005) Proceed ISB XX, p.102.Winter DA (1995) A.B.C. of Balance during Standing and

Walking. Waterloo: Waterloo Biomechanics.

Figure 2: Ensemble averages (±1 SD) of the eight muscles for one subject. Time normalized linear envelope EMG throughout gait initiation. EMGs order from top is lead-limb then trail-limb TA, ADD, TFL and ES.

T a b le 1 : M u sc le o n se t t im e s d u rin g g a it in itia tio n L -E S T -E S L -T F L T -T F L L -A D D T -A D D L -T A T -T A S u b je c t1 5 1 8 4 1 8 2 2 4 9 3 5 1 6 1 9 S u b je c t2 4 3 2 5 2 0 3 5 7 9 2 4 1 3 1 5 S u b je c t3 5 1 7 5 2 1 3 6 5 0 3 9 1 7 3 0 S u b je c t4 4 5 3 6 1 2 1 8 1 7 3 0 1 1 2 0 S u b je c t5 4 1 3 5 2 0 2 8 2 4 3 4 1 9 2 3 S u b je c t6 6 3 4 0 1 6 1 2 1 8 1 9 3 2 2 8 S u b je c t7 4 6 3 5 1 1 1 6 3 3 1 8 1 0 1 5 S u b je c t8 4 1 3 9 1 5 3 4 2 6 2 8 1 2 1 7 S u b je c t9 4 8 3 2 1 4 3 9 2 1 1 8 1 7 1 3 M e a n 4 7 .7 4 4 .6 1 6 .3 2 6 .7 3 5 .2 2 7 .2 1 6 .3 2 0 .0 S t D e v 7 .3 5 2 0 .4 3 .6 4 9 .9 4 2 0 .5 7 .9 2 6 .6 3 5 .9 4

T a b le 1 : M u sc le o n se t t im e s d u rin g g a it in itia tio n L -E S T -E S L -T F L T -T F L L -A D D T -A D D L -T A T -T A S u b je c t1 5 1 8 4 1 8 2 2 4 9 3 5 1 6 1 9 S u b je c t2 4 3 2 5 2 0 3 5 7 9 2 4 1 3 1 5 S u b je c t3 5 1 7 5 2 1 3 6 5 0 3 9 1 7 3 0 S u b je c t4 4 5 3 6 1 2 1 8 1 7 3 0 1 1 2 0 S u b je c t5 4 1 3 5 2 0 2 8 2 4 3 4 1 9 2 3 S u b je c t6 6 3 4 0 1 6 1 2 1 8 1 9 3 2 2 8 S u b je c t7 4 6 3 5 1 1 1 6 3 3 1 8 1 0 1 5 S u b je c t8 4 1 3 9 1 5 3 4 2 6 2 8 1 2 1 7 S u b je c t9 4 8 3 2 1 4 3 9 2 1 1 8 1 7 1 3 M e a n 4 7 .7 4 4 .6 1 6 .3 2 6 .7 3 5 .2 2 7 .2 1 6 .3 2 0 .0 S t D e v 7 .3 5 2 0 .4 3 .6 4 9 .9 4 2 0 .5 7 .9 2 6 .6 3 5 .9 4

Note: Muscle onset timings are all expressed as a percentage of total gait initiation (0-100%). L-ES & T-ES are the lead-limb and trail-limb erector spinae; L-TFL & T-TFL are the lead-limb and trail-limb tensor fasciae latae; L-ADD & T-ADD are the lead-limb and trail-limb adductor magnus; L-TA & T-TA are the lead-limb and trail-limb tibialis anterior.

Discussion

Presumably the two tibialis anterior muscles with the simultaneous release of the gastrocnemius/soleus muscles caused the posterior movement of the centre of pressure, whereas the lead-limb tensor fasciae latae (figure 2) contributed to the initial lateral shift toward the lead limb reported by Winter (1995). The delayed reaction of the erector spinae muscles confirm the kinetic analysis of gait initiation conducted by Robertson et al. (2005) that showed a brief period of falling prior to lead-limb heel-contact.

Discussion

Presumably the two tibialis anterior muscles with the simultaneous release of the gastrocnemius/soleus muscles caused the posterior movement of the centre of pressure, whereas the lead-limb tensor fasciae latae (figure 2) contributed to the initial lateral shift toward the lead limb reported by Winter (1995). The delayed reaction of the erector spinae muscles confirm the kinetic analysis of gait initiation conducted by Robertson et al. (2005) that showed a brief period of falling prior to lead-limb heel-contact.

Figure 1: From top to bottom: bilateral raw EMG of lead-limb then trail-limb TA, ADD, TFL, ES (blue) and vertical GRF data (black) for a single trial.