Cropped%20Logo%20-%20Background_edited.png

Patterson, C.W. (2010)

THE EFFECT OF CONTROLLING GAIT VELOCITY AND CADENCE ON THE GAIT SWING PHASE
A Pilot Study in Collegiate Walking

Human gait is a complex system that can be influenced by a multitude of external factors, including the natural aging process and certain pathologies. Despite these factors influencing both the used walking velocity and cadence of an individual, no research study has investigated how controlling these components are affecting the gait cycle. Instead previous research has focused on either the influence of walking velocity or cadence on the gait cycle, using restrictions within their research methodologies to assess the influences on a healthy population. The aim of this study was to investigate the effects of controlling both walking velocity and cadence on a healthy, young population. Using a combination of Brower timing gates and a nine-camera Vicon motion capture system with an integrated Kistler force platform, a range of commonly researched gait parameters in the gait’s swing phase was investigated in four university students gait. From the results, the author would suggest that future studies into the influences of pathology or aging on gait requires both a restriction on velocity and cadence for accurate assessment and comparison between population groupings.

An individual’s gait is affected by aging, injury and pathologies with these conditions influencing both their walking velocity and cadence (Whittle, 1996; Graf et al., 2005; Strike, 2009; Cofré et al., 2011). To gain full understanding of how the changes in walking velocity and cadence influence an individual gait, a frequently used methodological approach is to apply the velocity and/or cadence constrictions associated with the pathology or aging process to an unaffected population (Bohannon, 1997; Chen et al., 1997; Lelas et al., 2003; Graf et al., 2005; Cofré et al., 2011). However, the application of a constriction on both velocity and cadence parameters is a relatively unresolved issue.

Walking velocity is dependent on cadence and step length with shorter stride lengths resulting in higher cadences due to the decreased time that the limb is in its swing phase of the gait cycle (Selles et al., 2003; Cofré et al., 2011). Strike (2009) states the aim of the swing phase is to advance the limb unimpeded and the to prepare the limb for supporting the body weight for the next phase of the gait cycle. For sufficient ground clearance to occur the limb must be lifted from the ground and shortened primarily through the application of a propulsive ground reaction forces (GRFs) from ankle plantar-flexors, combined with flexion of both the hip and knee joints (Whittle, 1996; Neptune et al., 2001; Strike, 2009). The muscular actions of the limb in creating the kinematic motions necessary for the swing phase are produced by a series of proximal-to-distal power bursts (Winter, 1987; Lelas et al., 2003; Strike, 2009).

Firstly, the second power burst of the ankle joint (A2; Figure 1) occurs immediately prior to the initiation of the swing phase due to action of the ankle plantar-flexors to generate the necessary propulsive GRFs that push the limb up and forward to generate the required ground clearance (Whittle, 1996; Neptune et al., 2001; Strike, 2009). Lelas et al. (2003) has also stated the successful use of the A2 power burst as a predictor to both stride length and walking velocity. Final ground clearance is primarily controlled by the flexion of the knee to an estimated peak angle between 60° and 70°, an action created following the third power burst of the knee (K3; Figure 1). Not only does the K3 power burst increase ground clearance of the foot but the shortening of the limb lowers the moment of inertia in the lamb, making it easier to swing (Holt et al., 1991; Selles et al., 1999; Selles et al., 2003). The final component for the initiation of the swing phase is the flexion of the hip which pulls the limb further forward, this action is produced by the third hip power burst (H3; Figure 1). During terminal swing, the knee extends to prepare the limb for loading upon foot contact which is created by the final power burst at the knee joint (K4; Figure 1), with the velocity of this extension vital in determining stride length as increases in extension velocity bring the foot back to the ground quicker for loading.

Figure 1. Timing of sagittal plane power bursts for the ankle (A), knee (K) and hip (H) joints during the gait cycle (modified from Winter, 1987)

In studies where an individual’s walking velocity is controlled through the research question, alterations have been identified in a range of tempro-spatial parameters (Brinkmann & Perry, 1985; Enoka, 1994; Bohannan, 1997; Chen et al., 1997; Vardaxis et al., 1998; Mills & Barrett, 2001; Riley et al., 2001; Lelas et al., 2003; Selles et al., 2003; Damiano et al., 2004; Graf et al., 2005; Cofré et al., 2011). Normally, the key alterations are seen in either stride length, cadence or a combination of the two (Vardaxis et al., 1998: Graf et al., 2005) with further alterations seen in the time spent in the stance and double support phases of the gait cycle (Enoka, 1994; Vardaxis et al., 1998; Graf et al., 2005; Cofré et al., 2011). Studies have found positive correlations between walking velocities and the lower extremity joint power bursts (Chen et al., 1997; Vardaxis et al., 1998; Graf et al., 2005; Cofré et al., 2011) and with the kinematic actions that they generate. The positive correlations between gait kinematic actions and walking velocity include peak knee flexion and extension angles (Brinkmann & Perry, 1985; Bohanna, 1997; Chen et al., 1997; Mills & Barrett, 2001; Riley et al., 2001; Lelas et al., 2003; Selles et al., 2003; Damiano et al., 2004; Graf et al., 2005; Cofré et al., 2011) and the angular velocity of the knee extension.

Despite the knowledge that pathologies show altered velocity and cadence parameters, to this author’s knowledge, no study has investigated gait through controlling both walking velocity and cadence. In the current study, the selected walking velocities were within the stated normative ranges in Whittle (1996) with 1.0m/s and 1.3m/s commonly used as standardised walking velocities and are reported as self-selected slow and preferred velocities (Slider et al., 2008; Cofré et al., 2011). Cofré et al. (2011) found that when using the 1.0m/s and 1.3m/s velocities, participants reported cadences of 99 steps per minute and 113 steps per minute, respectively. The aim of this pilot study was to investigate the effect of controlling cadence on selected gait parameters when walking velocity has been set at a constant value. Firstly, selected propulsion characteristics responsible for initiating the gait’s swing phase will be investigated to ascertain if any change in these parameters occurs when stride length is altered through cadence control. Secondly, parameters from the swing phase were investigated to see if there are any significant differences during this phase with the associated restriction on both walking velocity and cadence.

Participant Information

A convenience sample of 4 university sports science students (3 male and 1 female; all aged 22 years old; height = 1.75 ± 0.09m; and mass = 76.20 ± 11.95kg; Table 1) volunteered for this pilot study. In keeping with the guidelines outlined by an independent ethics board at the University of Roehampton, the participants were given a full briefing on the study, including the methodology, and their roles within it before providing the researchers with a signed consent form. The briefing occurred during a familiarisation period for the participants to become accustomed to the desired walking velocities and cadences required for both experimental conditions as well as the use of the 3D motion capture equipment and the data collection methodology.

Table 1. Participant's anthropometric data

Laboratory Set-Up

All data collection, processing and analysis took place in the biomechanics laboratory at the University of Roehampton, United Kingdom . At the end of a 10m runway inside the laboratory environment was a Vicon motion capture system consisting of nine infra-red cameras (Vicon MXF20; Vicon Motion Systems & Peak Performance Inc.; Oxford, UK) calibrated to an error rate of <0.01m and capturing at 100Hz sampling rate (Figure 2). Within the capture volume of the camera system was two imbedded Kistler force platforms (Kistler Model 9281C; Kistler Instruments Ltd.; Hampshire, UK) which were synchronised with the motion capture system at a sampling rate of 1000Hz, both of which were zeroed immediately prior to each participant’s first trial (Figure 2). Two pairs of Brower timing gates were positioned 2m apart laterally alongside the force platforms with the infra-red beam emitters (Brower IRD-T175 Timing Gate Detector; Brower Timing Systems; Draper, UT, USA) and detectors (Brower IRE Timing Gate Emitter; Brower Timing Systems; Draper, UT, USA) placed 1m apart either side of the participant’s direction of movement along the runway and force platform (Figure 2). To indicate required cadences during the trials, a metronome (Steinway Metronome iPhone Application Version 1.1.3.; Steinway Musical Instruments Inc.; Waltham, MA, USA) was placed on a table alongside the testing area, but still outside of the capture volume of the cameras.

Figure 2. Laboratory schematic and equipment locations for data collection

Sixteen 10mm retro-reflective markers were attached, using double-sided medical tape, to the participant’s lower extremities in specific anatomical locations to create the kinematic model for processing and analysis (Figure 3). Four markers placed on both anterior superior iliac spinae (LASI and RSAI) and posterior superior iliac spinae (LPSI and RPSI) were used to construct the model’s pelvis (Figure 3). Markers were attached to the lateral femoral epicondyles (LKNE and RKNE) and lateral malleoli (LANK and RANK) to identify the joint centres of the knee and ankle, respectively (Figure 3). The feet were denoted by a marker placed on top of the foot on the 2nd metatarsal head (LTOE and RTOE) and on the calcaneous (LHEE and RHEE) at the same height off the floor as the toe markers (Figure 3).

Figure 3. Anatomical locations used for retro-reflective marker placements

Variable Definition

Walking velocity was calculated by dividing the 2m distance between the pairs of timing gates by the time taken in seconds for the participant to cover the distance. Cadence was defined as the number of steps taken per minute by the participant while walking within the capture zone of the motion capture system. Stride length was measured as the distance between two consecutive heel strikes of the same foot during the strides over the force platforms. The stance phase of the stride was identified by the duration between the identified moments of heel strike and toe-off points while the time between the toe-off point in one stride to the heel strike of the next was used to define the swing phase of the walking stride. Both variables were expressed as a percentage of the overall duration of the stride with all tempro-spatial parameters collected from the participant’s preferred limb. For kinematic analysis, a trio of knee kinematic parameters were measured, namely peak knee flexion and extension angles during swing phase, and the velocity of knee extension at the end of the swing phase. Peak knee flexion was identified as the peak angle at any moment during the swing phase whereas peak knee extension was identified as the joint’s extension angle at the heel strike of the following stride. The velocity at which the knee extended to its peak extension angle from the peak flexion angle was acknowledged as the extension velocity parameter. For all power analysis, the A2, H3, K3, and L4 sagittal plane power bursts were selected following their definitions from previous research (Winter, 1987). Each power burst was expressed as the peak magnitude of the generation or absorption power burst recorded and was then normalised by the participant’s mass for between-participant and between-study comparisons.

Data Collection Procedure

Except for the knee and ankle joint widths, which were collected prior to retro-reflective marker placement, all anthropometric data was collected for the participant’s 3D kinematic model prior to the short familiarisation period and after the attachment of the retro-reflective markers. Participant’s height and mass were recorded using a Seca stadiometer (Seca Leicester Portable Stadiometer; Seca; Hamburg, Germany) and CMS weighting scales (CMS weighing scales; CMS Weighing Equipment Ltd.; London, UK) , respectively, with the joint widths measured by a 14cm bicondylar calliper (Holtain 14cm Bicondylar Calliper; Holtain Ltd.; Crymych, UK) between the medial and lateral femoral epicondyles and between the medial and lateral malleoli for the knee and ankle measurements. A standard fabric measuring tape that denoted centimetre distances was used to measure the distance between the medial malleolus and the ASIS marker of the participant’s preferred limb, a distance used to define leg length. Following the participant’s briefing, marker attachment, anthropometric data collection, a short familiarisation session for the participant to become accustomed to the equipment and testing protocols, and the signing of participation consent forms the data collection could commence. The participant found an appropriate starting point along the available 10m runway that would result in a successful foot strikes on both the force platforms. Five successful trials were recorded for each participant and each of the four-experimental conditions (Table 2), the order of the experimental conditions determined through a randomised draw prior to that condition’s data collection. For a trial to be successful, clear and clean foot strikes were recorded on the force platforms without the participant “targeting” the platforms while matching the desired velocity and cadences (Table 2).

Table 2. Experimental conditions outlining the desired walking velocity and cadence

Data Processing and Analysis

The raw data was captured and processed through Vicon Nexus software10 for the 3D lower extremity model to be created for each participant. All raw data was filtered using a Woltring mean squared error (MSE) filtering routine with a pre-set noise value of 15mm. Using a combination of GRF data and the autocorrelate function in Nexus, the heel strike and toe-off events were identified with a 20N GRF threshold used to identify each event. Additionally, the use of the function allowed isolation of the swing phase for analysis. While joint kinematic and power bursts were processed in the Nexus software, the tempro-spatial parameters required the use of Vicon Polygon software11 for processing. The tempro-spatial parameters require two recorded gait cycles with the feet in contact with the force platforms, the power bursts and kinematic parameters were assessed from one gait cycle. All data was exported to Microsoft Excel12 for analysis and presentation.

All parameters showed an increase with the increase in walking velocity (Table 3). Neither recorded cadence at each condition was at the desired value when the restriction was applied, however the controlled cadence at the slower walking velocity was within the acceptable ±2% allowance (Table 3). When the cadence restriction was applied to both walking velocity conditions, all but two parameters decreased (Table 3). The parameters that increased when the cadence was, in attempt, constricted were stride length and peak knee extension angle (Table 3).

Table 3. Comparison of results for the different walking velocities with and without the cadence control (standard deviations in parenthesis)

The aim of this pilot work was to investigate the effect of applying cadence control on gait parameters at two specific walking velocities. One of the velocities was like those reported for a similarly aged population to the sample in the study while the other was seen in an aging population or one suffering from certain pathologies (Whittle, 1996; Bohannon, 1997) which would represent a constraint on the study’s population sample. All parameters showed a difference between the two velocities with differences also present within the velocity conditions when a cadence was controlled. To this author’s knowledge, few studies have investigated the use of a cadence restriction while instead favouring for the use of a walking velocity restriction and none have utilised both a walking velocity and cadence restriction. In agreement with previous literature, increases in walking velocity are the result of a combined alteration of both cadence and step length (Vardaxis et al., 1998; Riley et al., 2001; Graf et al., 2005; Cofré et al., 2011). As walking velocity increased, the joint power bursts increased as expected which explains the increase in kinematic parameters found to be correlated with walking velocity (Chen et al., 1997; Riley et al., 2001; Graf et al., 2005; Cofré et al., 2011).

Walking velocity is created through the combination of cadence and stride length therefore when applying a restriction to the cadence capability, the maintenance of velocity becomes the responsibility of the influencing factors on stride length. Although not directly researched, the resulting changes in cadence have been reported and explained by previous gait research (Kerrigan et al., 1998; Vardaxis et al., 1998; Riley et al., 2001; Graf et al., 2005; Lewis & Ferris, 2008; Cofré et al., 2011). The decreases in joint power bursts seen within the controlled cadence conditions have been identified as they key influencing factor in the alteration of stride length (Graf et al., 2005; Lewis & Ferris, 2008; Cofré et al., 2011). The findings of this study, further highlight the complex relationships between interacting gait parameters.

The combination of the lessened A2S, K3S and H3S bursts lead to a decrease in the generated propulsive GRFs and the flexion of both the hip and knee joints (Whittle, 1996; Vardaxis et al., 1998; Neptune et al., 2001; Lewis & Francis, 2008; Strike, 2009). The lack of propulsive forces and joint flexions result in lower ground clearance while lengthening the leg thus increases the inertial properties of the swinging limb (Holt et al., 1991; Selles et al., 1999; Selles et al., 2003; Strike, 2009). The outcome of this combination is the decreased duration of the swing phase despite the longer stride length (Neptune et al., 2001; Selles et al., 2003; Cofré et al., 2011). The other factor in the lengthening of the stride length is the decreased K4S power burst, the burst which is responsible for the extension of the knee joint (Winter, 1987) with this decreased power burst mirrored in the decrease in the velocity of the knee extension motion. The slower extension prolongs the terminal stage of the swing phase by keeping the foot off the ground for a longer time, resulting in a longer stride length which, hypothetically, is characterised by reaching further forward with the planting foot towards the end of the swing phase.

The findings of this pilot study must be taken in consideration given the small population sample used in the study, although the findings are also questionable in their validity as from further analysis the cadence utilised in the fast walking trials was outside of the acceptable 2% allowance. The findings of one participant can not only be attributed to causing the questionable cadence recordings but also highlight the link between the length of an individual’s legs and their walking gait mechanics (Kaufman et al., 1996; Gurney, 2002). The participant in question had a shorter leg length resulting in a shorter stride length which ultimately meant a higher cadence to match the velocities of the other participants (Table 1). Further hypothesised explanations for the misleading findings include the use of a subjective audio-based assessment of cadence and the inability to confirm the cadences prior to the processing procedures. In addition to addressing the potential limitations, further research would suggest continuing the line of the research question using a larger population sample, if not also different population samples.

Bohannon, R.W. (1997). Comfortable and maximum walking speed of adults aged 20-79 years: Reference values and determinants. Age and Aging, 26: 15-19.
Brinkmann, J.R., and Perry, J. (1985). Rate and range of knee motion during ambulation in healthy and arthritic subjects. Physical Therapy, 65(7): 1055-1060.
Chen, I.H., Kuo, K.N., and Andriacchi, T.P. (1997). The influence of walking speed on mechanical joint power during gait. Gait & Posture, 6: 171-176.
Cofré, L.E., Lythgo, N., Morgan, D., and Galea, M.P. (2011). Aging modifies joint power and work when gait speeds are matched. Gait & Posture, 33(3): 484-489.
Damiano, D.L., Laws, E., Carmines, D.V., and Abel, M.F. (2006). Relationship of spasticity of knee angular velocity and motion during gait in cerebral palsy. Gait & Posture, 23: 1-8.
Enoka, R.M. (1994). Neuromechanical Basis for Kinesiology 2nd Edition. Champaign, IL: Human Kinetics.
Graf, A., Judge, J.O., Ounpuu, S., and Thelen, D.G. (2005). The effect of walking speed on lower extremity joint powers among elderly adults who exhibit low physical performance. Archive of Physical Medicine and Rehabilitation, 86: 2177-2183.
Gurney, B. (2002). Leg length discrepancy. Gait & Posture, 15(2): 195-206.
Holt, K.G., Hammill, J., and Andres, R.O. (1991). Predicting the minimal energy costs of human walking. Medicine and Science in Sports and Exercise, 23: 491-498.
Kaufman, K.R., Miller, L.S., and Sutherland, D.H. (1996). Gait asymmetry in patients with limb-length inequality. Journal of Paediatric Orthopaedics, 16(2): 144-150.
Lelas, J.L., Merriman, G.J., Riley, P.O., and Kerrigan, D.C. (2003). Predicting peak kinematic and kinetic parameters from gait speed. Gait & Posture, 17: 106-112.
Lewis, C.L., and Ferris, D.P. (2008). Walking with increased ankle push-off decreases hip muscle moments. Journal of Biomechanics, 41: 2082-2089.
Mills, P.M., and Barrett, R.S. (2001). Swing phase mechanics of healthy young and elderly men. Human Movement Science, 20: 427-446.
Neptune, R.R., Kautz, S.A., and Zajaz, F.E. (2001). Contributions of the individual ankle plantar flexors to support forward progression and swing initiation during walking. Journal of Biomechanics, 34: 1387-1398.
Riley, P.O., Della-Croce, U., and Kerrigan, D.C. (2001). Effect of age on lower extremity joint moment contributions to gait speed. Gait & Posture, 14: 264-270.
Selles, R.W., Bussmann, J.B., Wagenaar, R.C., and Stam, H.J. (1999). Effects of prosthetic mass and mass distribution on kinematics and energetics of prosthetic gait: A systematic review. Archive of Physical Medicine and Rehabilitation, 80: 1593-1599.
Selles, R.W., Korteland, S., Van Soest, A.J., Bussman, J.B., and Stam, H.J. (2003). Lower-leg inertial properties in transtibial amputees and control subjects and their influence on swing phase during gait. Archive of Physical Medicine and Rehabilitation, 84: 569-577.
Slider, A., Heiderscheit, B., and Thelen, D. (2008). Active and passive contributions to joint kinetics during walking in older adults. Journal of Biomechanics, 41: 1520-1527.
Strike, S. (2009). Gait and ergonomics: Normal and pathological. In: Ergonomics for Rehabilitation Professionals (edited by S.Kumar), pp. 137-171. Florida, USA: Taylor and Francis Group.
Vardaxis, V.G., Allard, P., Lachance, R., and Duhaime, M. (1998). Classification of able-bodied gait using 3D muscle powers. Human Movement Science, 17: 121-136.
Whittle, M.W. (1996). Gait Analysis: An Introduction 4th Edition. Philadelphia, USA: Elsevier.
Winter, D.A. (1987). The Biomechanics and Motor Control of Human Gait. Waterloo, Ontario: University of Waterloo.

THE EFFECT OF CONTROLLING GAIT VELOCITY AND CADENCE ON THE GAIT SWING PHASE A Pilot Study in Collegiate Walking

THE EFFECT OF CONTROLLING GAIT VELOCITY AND CADENCE ON THE GAIT SWING PHASE
A Pilot Study in Collegiate Walking