 |
 |
 |
 |
SPATIAL-TEMPORAL SYMMETRY OF THE LOWER LIMBS DURING NORMAL GAIT
Dr. Ch. CHRISTOFORIDIS., Dr. A. KAMPAS., Dr. K. MARTINIDIS,
Dr. N. AGGELOUSSIS, Dr. I. FATOUROS., Dr. N. VEZOS,
Dr. K. TAXILDARIS., Dr. G, MAVROMATIS
Democritus University of Thrace
Department of Physical Education and Sport Science
Keywords: gait, natural speed, linear kinematics, symmetry, biomechanics
Introduction
Gait analysis is a method that is commonly used in order to survey disorders of the legs. It is based on the principle that walking is the most natural movement of the human body and it requires the least energy consumption of all movements. Consequently it might be the only movement that for which a normal movement model can be defined (Allard et al, 1996). By the gait analysis method, the data of persons with movement disorders at their legs are being compared with normal walking models and by this way we find out the deviations of the normal pattern of walking so we can do a diagnosis of the reasons that causes these deviations. Because of that fact the validity of that method depends on the validity of the walking normal pattern's that is used as a standard (Sadeghi et al, 2000).
Among the important parameters that we should be aware of in order to create the model of the normal walking is the presence or the absence of symmetry on the movement of the left and right legs during walking and the elements that affect it. Using new technology we are able to define clearly the differences between the legs in kinematics and other elements of walking (Allard et al, 1996). Findings of previous researches don't completely give a solution to that problem (Craik et al, 1985). In some researches there are not significant differences found between the legs in time or kinematic parameters of young adults (Allard et al, 1996), while in others is mentioned that there are significant differences between the legs (Sadeghi et al, 2000).
According to Winter (1995), there is an asymmetry between some mechanical parameters of normal gait caused by the difference of the length of the legs, by the domination and the asymmetry of brain hemispheres' maturation. In particular about brain's function during walking, Gabell & Navak (1984) supported the idea that the length and the duration of the stride are controlled by the system that is responsible for the general pattern of walking while the width of stride and the duration of stance using both legs are controlled by the balance control system.
Newer researchers' results showed that the asymmetries that are observed between legs in walking parameters are caused by the different tasks that each leg undertakes during walking, such as to provide stability and to move the body forward (Sadeghi et al, 2000). There is not a research yet for the Greek population in particular about the asymmetries of kinematic characteristics of the legs during walking. This information is important in order to create a model of normal walking that refers to the Greek population which are valid and can be used to make a diagnosis of disorders of the legs by analyzing the walking pattern of persons of the same population.
The present study was aimed at tracing the differences between the lower limbs concerning kinematic parameters during the normal gait, so as to draw conclusions about the existence or not of asymmetries and about the functional state of the lower limbs during gait.
Methods
Sample and Procees
The sample of the research consisted of 60 male students of the Faculty of Sports Science and Physical Education of Democritus University of Thrace, Greece. Their age was 23.2 3.2 years, their height: 1.82 0.05m and their body mass: 83.3 7.9 kg, with no muscle-skeletal problems at the lower limbs.
The movement studied was walking at natural speed. Before commencing to record the data, the width of natural walking speed of each participant was calculated. The term "natural walking speed" refers to the speed that each individual chooses to walk at. Initially, a timing system of photocells was used to determine the time needed by the participant to cover a distance of 6.3m at walking pace. This distance was chosen to allow the participants to walk enough steps, in order to have a more reliable estimation of their natural walking speed.
The photocells were placed at the height of the shoulders of the participants, so as to avoid their false activation due to arm or leg movement (Herzog et al, 1989). The participants executed five repetitions (Hamill et al, 1985), on a 10m long carpet. Consequently, the natural walking speed width of each participant was determined through the following equation:
with: X the mean value of the five repetitions, t the critical value for a significance level of 0.005 and Ν-1 degrees of freedom, S the standard deviation of the five repetitions Ν the number of repetitions. Any walking speed value fell into the range of the space calculated through the equation above, was considered as natural walking speed. The kinematical analysis of walking was done by using 3-dimensional system consisted of two cameras with a sampling frequency of 60 Hz. Cameras where placed in a way so their optic axons made a 90 degrees angle. Synchronization of the cameras was done by using an electronic circuit of switchers and LEDs. The synchronization point of the two cameras was the appearance of the light of the LED, which was placed on the heel of the tested stance foot, while it was in the first contact with the ground. Before the recording of the movement, 4 reflecting markers were placed on the left and right side of the body: at the upper phalange of big toe, at the tuber of 5th metatarsus' base, at the end of outside surface of the heel and at the end of inside surface of the opposite heel.
Walking area gradated using a cube with 16 checkpoints. Cube's dimensions were 180x90x180cm and corresponded to experiment's conducting area (Ladin, 1995). Participants walked with their natural speed five times from each direction in order to record both legs' movement. The closest to average walking speed was chosen from the total of five attempts. By using the Ariel Performance Analysis System (APAS), one walking cycle was analyzed. Kinematic data processing included the digitization of videotapes, the transformation of 2-dimensional coordinates into real 3-dimensional values by using the method of direct linear transformation (DLT).
For the purpose of the analysis, the gait cycle was divided into the following events (figure 1): 1st contact of the heel of the examined foot with the ground (a), foot flat, i.e. full contact of the sole of the examined foot with the ground (b), opposite foot toe-off from the ground (c), examined foot heel-off from the ground (d), opposite foot heel contact with the ground (e), toe-off of the examined foot from the ground (f), foot clearance i.e. toes of the examined foot which swings next to the toes of the opposite foot (g), the shank of the swinging examined foot comes vertically to the ground (h) and the 2nd heel contact of the examined foot with the ground (i).
According to the events mentioned above, the gait cycle was divided in the phases below and the time needed for each phase was calculated as a percentage of the total gait cycle time: initial double stance (from the 1st contact of examined heel with the ground until the opposite toe-off), single stance (from examined foot toe-off until the opposite foot heel contact), final double stance (from opposite foot heel contact until the examined foot toe-off), total stance (from the 1st contact of examined heel with the ground until it's toe-off), initial swing (from examined foot toe-off until foot clearance), middle swing (from foot clearance until the shank of the examined foot is vertical to the ground), final swing (from the moment that the shank of the swinging foot is vertical to the ground until its 2nd heel contact with the ground) and total swing (from the examined foot toe-off until it's 2nd heel contact).
 |
Moreover, it was calculated the duration of total double stance (as the sum of initial and final double stance), the duration of stride time (sec) and the quotient of the double stance to the single stance.
The linear kinematic characteristics that calculated in meters, except step angle which calculated in degrees, were (figure 2): step length (a), stride length (c), step width (b), stride width (d), the base of support (e) and the step angle (f).
In order to check the reproducibility of the movement, all the attempts of one person were analyzed for the left and the right foot. Then, the intra-class correlation factor of each foot was calculated, and its values were 0.85 and 0.82 correspondingly for the right and the left foot, which were absolutely acceptable. Moreover, in order to check the reliability of the analysis, the gait of three people was analyzed twice and the intra-class correlation factors between the repeated analyses were calculated. The values of the factors varied between 0.90 and 0.93.
 |
Results
In order to check the variance between the legs' in the variables mentioned above during walking, a Paired T-test was done with a level of significance assigned at .05. According to the results there were statistically significant differences between the left and right leg in the time, expressed as percentage of gait cycle, that took place the following temporal events (Table 1): full contact of the sole of the examined foot (t60 = -2.15, p<.05), examined foot heel-off (t60 = -5.90, p<.001), opposite foot heel contact (t60 = -9.96, p<.001), foot clearance (t60 = -4.95, p<.001), and swinging shank vertical to the ground (t60 = -4.95, p<.001). On the contrary, there were not statistically important differences found between left and right leg in time that the following events occurred: opposite foot toe-off (t60 = -1.46, p = .15) and examined foot toe-off (t60 = 0.32, p = .75).
Table 1
Means and standard deviations of the time that took place the temporal events of the gait cycle (expressed as % of the gait cycle) for the two limbs
|
||||||||||||||||||||||||||||||||||||
As far as concerned the phases of the gait cycle it was found that there were statistically significant differences between the lower limbs (Table 2) in duration, expressed as percentage (%) of the stride, of the: single stance (t60= -6.81, p<.001), final double stance (t60= 10.59, p<.001), initial swing (t60= -4.87, p<.001), final swing (t60=3.52, p<.001), total double stance (t60= 8.95, p<.001), stride time (t60=3.5, p<.001) and the quotient double to single support (t60= 9.99, p<.001). On the contrary, there were not found statistically significant differences between the right and the left lower limb in duration, expressed as percentage (%) of the stride, of the: initial double stance (t60=-1.46, p=.15), total stance (t60=0.32, p=.75), middle swing (t60=1.22, p=.23) and total swing (t60=-0.32, p=.75).
About the linear kinematic characteristics of the stride, statistically significant differences were found, between the lower limbs (Table 3) in: step length (t60 = -5.1,p<.001), stride length (t60=-3.84, p<.001), step width (t60=2.56, p<05), stride width (t60=3.396, p<.01) and in base of the support (t60=-3.11, p<.01). On the contrary, no statistically significant differences were found, between the right and the left lower limb, in step angle (t60=-1.16, p=.25).
Table 2
Means and standard deviations of the gait cycle phase's duration (expressed as percentage of the gait cycle), stride time (sec) and the quotient double/single limb support time
|
||||||||||||||||||||||||||||||||||||||||||||||||||||
Table 3
Means and standard deviations of the linear kinematic characteristics
|
||||||||||||||||||||||||||||||||
Discussion
In this paper the natural gait of Greek adults was analyzed in order to find, if exist; the significant differences between the lower limbs on the spatial-temporal characteristics of gait. It was found from the results that there were statistically significant differences between the lower limbs, almost to the whole of the temporal parameters of the walking cycle (six from the eight events and eight from the 12 phases). The model of disproportion was common to all time characteristics which were significantly different between the two strides, with the incidents of the right stride to appear earlier with regard to the incidents of the left stride. As for the durations of the phases goes, greater duration of single support and initial swing were noticed in the left stride, while the rest phases – about which there were significant differences between the two strides – have had greater duration in the right stride. The above findings are in accordance with the results of previous studies (Gundersen et al, 1989; Wheelwright and et al, 1993), which found out significant differences between the lower limbs, in the time parameters of walking.
On the contrary, in the studies carried out by Baker & Hewison (1990), Allard et al. (1996), Whittle (1996) and Sadeghi et al. (1997), no significant differences between the lower limbs in the time characteristics of walking were found out. Also, no significant differences have been mentioned between the right and the left lower limbs, in healthy adults (Konczak, 1994; Jaegers et al, 1995), but in these studies the walking cycle is only divided into the support and swing phase.
The lack of significant differences between the lower limbs in the time that the facts of the gait cycle occurred and in the durations of the phases of the gait cycle which was found out on the above studies is probably due to the division of the walking cycle to a small number of facts and phases. In the present study, not only the big size of the sample but the division of the walking cycle into a big number of facts and phases as well, confirm the validity of the results, with regard to the existence of asymmetry in the temporal characteristics of normal walking.
With regard to the basic kinematic characteristics, it was found out that the two lower limbs were significantly different in all parameters except for the step angle. In accordance with the findings of the present study are the findings of Gundersen et al. (1989) and Wheelwright et al. (1993), who concluded in a non-symmetrical model of behaviour of the lower limbs during normal walking, in the movement characteristics of walking.
As far as the step length is concerned, lack in differences between the lower limbs in healthy adults, was reported in the studies of Vaughan et al. (1992), Prince et al. (1993), Craik & Oatis (1995) and Allard et al. (1996). With regard to the stride length no significant differences were reported in the researches of Vaughan et al. (1992), Prince et al. (1993), Zhu et al. (1993), Craik & Oatis (1995), Jaegers et al. (1995) and Allard et al. (1996). Finally, about step angle, Feltner et al. (1994) noticed differences in the step angle between the two lower limbs, which was noticed in the present research as well.
It is worth mentioning that although the above researchers accept the existence of differences in some parameters of walking, they claim that the behaviour of a dynamic system, such as human body, can't be described as asymmetrical on the whole, because there are statistically significant differences, between the right and the left lower limb, in some variables. These researchers compare the power of the statistic analysis with the strength of the adaptable behaviour of the lower limbs in the conditions given, which is called natural functional asymmetry of walking (Witer & Sienko, 1988; Sadeghi et al, 2000).
In the researches where the model of the spatial and temporal characteristics of walking is symmetric, the small number of the sample tested may constitute a restrictive factor as for the statistic significance of the differences of the lower limbs. On the contrary, the statistic significance of the differences of the lower limbs which was found out almost on the whole of the kinematic parameters, in this survey, is probably due to the large number and the great homogeneity of the sample, as far as the height and the length of the lower limbs are concerned, characteristics which affect significantly the basic kinematic parameters of walking. So, although the movement variable means had small differences between the lower limbs, the very small deviations of the variables from the mean value, to all subjects of the sample, resulted in these differences to be statistically significant.
What's more, according to Winter (1991) the asymmetry between the two lower limbs is regarded natural in a way and even necessary for the purposes of the movement to be achieved. So, the differences found in this research, may be due to the different obligations of the two lower limbs as far as the support and the propulsion of the body are concerned during walking. In particular, a need to spend more time on the limb which is responsible for the support of the body and the maintenance of the balance is evident, while it is not vital that the same time be spent on the limb which is responsible for the propulsion of the body (Sadeghi et al. 1997) so, the limb which is responsible for the propulsion of the body may make smaller and faster steps, with regard to limb which is responsible for the support of the body, so that the best possible result can be achieved, as far as the function of the movement is concerned (Winter 1991). Another possible reason for the disproportion of the lower limbs during walking is the existence of dominance in the lower limbs (Hirokawa, 1989).
Finally, it is worth mentioning that despite the differences in the isolated parameters tested, the total temporal model of walking, as reflected from the durations of the total support and total swing is not different between the two lower limbs. As a result, the natural model of the strides was not disturbed and there was a harmonious time flow in the movement. So, it could be concluded that the differences in the parameters of walking tested do not affect the harmonious flow of the whole movement and consist the person's muscle-skeletal system adaptation to every natural and anatomic factors affecting the function of the limbs participating in the movements.
Abstract
The purpose of the current research was the comparative study of the kinematic characteristics of the lower limbs during normal gait. The subjects were sixty students without muscle-skeletal problems in the lower limbs. Walking was studied at the subjects' natural speeds, which were measured by an electronic system consisted of two photocells and a time-recorder. Then, three-dimensional recording of each subject's gait was performed by two video cameras that were placed in the same side of a 10m long walkway. An electrical circuit involved LEDs was used for the synchronization of the two cameras. The frames of the movement were digitized using the Ariel Performance Analysis System (APAS) and the coordinates of selected points on the body were transformed to original values using the DLT procedure with 16 control points. Low-pass digital filters were used to smooth the raw data with cut-off frequencies 1-6Hz. Then, temporal and linear kinematic data were calculated for each lower limb on the sagittal and frontal plane, during one right and one left gait cycle. Paired T-tests were used for the statistical treatment of the data.
The results showed significant differences between the two lower limbs in most of the temporal events (p<.001 in four events and p<.05 in one event), the temporal phases (single stance: t60= -6.81, p<.001, final double stance: t60= 10.59, p<.001, initial swing: t60= -4.87, p<.001, final swing: t60=3.52, p<.001), total double stance t60= 8.95, p<.001, stride time: t60=3.5, p<.001 and the quotient double to single support: t60= 9.99, p<.001) and the linear kinematic variables (step length: t60 = -5.1,p<.001, stride length: t60=-3.84, p<.001, step width: t60=2.56, p<05, stride width: t60=3.396, p<.01 and base of support: t60=-3.11, p<.01). These findings supported the statement that natural gait is not perfectly symmetrical.
References
ALLARD, P., LACHANCE, R., AISSAOUI, R. AND DUHAIME, M., Simultaneous bilateral 3-D gait analysis. In: Human Movement Science. 15, 327-346, 1996
BAKER, P.A. AND HEWISON, S.R, Fall injuries in the elderly. In: Clinics in Geriatrical Medicine. 1, 501-512. 1990
CRAIK, R.B.L., LEIPER, C., COZZENS, B.A., ROAPER, T. AND HANLON, M, Comparisons of spatial and temporal patterns in walking. In: International Series on Biomechanics. Champaign, Illinois. Volume 5, A. Human Kinetics Publishers, 555-559, 1985
CRAIK, R.B.L. AND OATIS, C.A., Spatial and temporal characteristics of foot fall patterns. In: Gait Analysis: Theory and Application. CRAIK, R.B.L. & OATIS, C.A. (Eds). St Louis, MN. Mosby, 143-158. 1995
FELTNER, M.E., MACRAE, P.G. AND MCNITT - GRAY, J.L., Quantitative gait assessment as a predictor of prospective and retrospective falls in community-dwelling older women. In: Archives of Physical Medicine and Rehabilitation. 75, 447-453, 1994
GABELL, A. AND NAYAK, U.S.L., The effect of age on variability in gait. In: Journal of Gerontology. 39, 662-666, 1984
GUNDERSEN, L.A., VALLE, D.R., DANOFF, J.V., STANHOPE, S.J. AND SNYDER-MACKLER, L., Bilateral analysis of the knee and ankle during gait: an examination of the relationship between lateral dominance and symmetry. In: Physical Therapy. 69, 640-650, 1989
HIROKAWA, S., Normal gait characteristics under temporal and distance constraints. In: Journal of Biomedical Engineering. 11 (6), 449-456, 1989
JAEGERS, S.M.H.J., ARENDZEN, J.H. AND DEJONGH, H.J., Prosthetic gait of unilateral transfemoral amputees: a kinematic study. In: Archives of Physical Medicine and Rehabilitation. 76, 736-743, 1995
KONCZAK, J., Effects of optic flow on the kinematics of human gait: a comparison of young and older adults. In: Journal of Motor Behavior. 26 (3), 225-236, 1994
LADIN, Z., Three-Dimensional Instrumentation. In: ALLARD, P., STOKES, I.A.F. & BLANCHI, J.P. (Eds). Three-Dimensional Analysis of Human Movement. Champaign, Illinois. Human Kinetics Publishers. 3-17, 1995
PRINCE, F., ALLARD, P., MCFADYEN, B.J. AND AISSAOUI, R., Comparison of gait between young adults fitted with the Space foot and nondisabled persons. In: Archives of Physical Medicine and Rehabilitation. 74, 1369-1376, 1993
ROSE, J. AND GAMBLE, J.G. In: Human walking. 2nd Edition. Williams & Wilkins. New York. Waverly Company, 1994
SADEGHI, H., ALLARD, P. AND DUHAIME, M., Symmetry and limb dominance in able-bodied subjects. In: Human Movement Science. 16, 243-258, 1997
SADEGHI, H., ALLARD, P., PRINCE, F. AND LABELLE, H., Functional gait asymmetry in able-bodied gait: a review. In: Gait & Posture. 12, 34-45, 2000
VAUGHAN, L.C., DAVIS, B.L. AND O'CONNOR, J.C., In: Dynamics of Human Gait. Champaign, Illinois. Human Kinetics Publishers, 1992
WHEELWRIGHT, E.F., MINNS, R.A., LAW, H.T. AND ELTON, R.A., Temporal and spatial parameters of gait in children 1: Normal control data. In: Developmental Medicine and Child Neurology. 35 (2), 102-113, 1993.
WHITTLE, M.W., In: Gait Analysis: an introduction. Jordan Hill, Oxford: Butterworth–Heinemann, Biddles Ltd, 1996
WINTER, D.A. AND SIENKO, S.E., Biomechanics of below-knee amputee gait. In: Journal of Biomechanics. 21, 361-367, 1988
WINTER, D.A., In: Biomechanics and Motor Control of Human Movement. 2nd Edition. . New York: John Willey & Sons, 1990
WINTER, D.A., In: Biomechanics and Motor Control of Human Gait: Normal, Elderly and Pathological. 2nd Edition. Waterloo, Ontario, Canada: University of Waterloo, 1991
WINTER, D.A., Human balance and posture control during standing and walking. In: Gait & Posture. 3, 193-214, 1995
ZHU, H., WERTSCH, J.J., HARRIS, G.F., ALBA, H.M. AND PRICE, M.B., Sensate and insensate in-shoe plantar pressures. In: Archives of Physical Medicine and Rehabilitation. 74, 1362-1368, 1993
