Biology of Sport, Vol. 21 No3, 2004
.
COMPARISON OF ASYMMETRIES IN GROUND REACTION FORCE
PATTERNS BETWEEN NORMAL HUMAN GAIT AND FOOTBALL
PLAYERS
B.S.Cigali1, E.Ulucam1, A.Yılmaz1, M.Cakıroglu2
1
Dept. of Anatomy, Faculty of Medicine, Trakya University, Edirne, Turkey;
2
School of Sports Sciences, Trakya University, Edirne, Turkey
Abstract. The purpose of this study is to investigate the asymmetry of some
temporal gait parameters and Ground Reaction Forces (GRF) in normal subjects
and football players. 31 football players and 33 normal subjects participated in our
study. The gait parameters were recorded by using an insole system. GRF values
were recorded from the heel (Fmax1), middle feet (Fmax2), forefeet lateral side
(Fmax3) and forfeet medial side (Fmax4). The subjects were asked to walk along
an 8 m footpath and time versus force graphics recorded. Data were collected after
exporting a worksheet program for percentage of swing and stance phase time
(Tswing, Tstance), double support time (DST), and GRF values from four
different parts of their feet and times to reach maximum force values (Tmax1,
Tmax, Tmax3, Tmax4). Temporal parameters of both groups Tswing, T stance and
DST have no statistical differences but they have no exact symmetry as well. All
Fmax values were significantly high for the left side of the football players and in
the control group only Fmax1and Fmax4 were significantly high for the left side.
Moreover, when comparing both groups, while for the left side Fmax3 and Tmax4
were significantly high for the football players group, Fmax1 was significantly
low. For the right side only Fmax3 was significantly high for the football players.
In conclusion, GRF values show asymmetry especially for the football players
because of their stronger muscles coupled with the fact that they could stop and
propel themselves into motion better than the control group.
(Biol.Sport 21:241-248, 2004)
Key words: Football player - Asymmetry - Ground reaction force - Gait - Human
locomotion - Force distribution
-
-
-
-
-
Reprint request to: Assist. Prof. Bülent Sabri Cigali, Dept. of Anatomy, Faculty of
Medicine, Trakya University, 22030 Edirne Turkey
Tel./fax: +90 284 2355935; E-mail: [email protected]
242
B.S.Cigali et al.
Biol.Sport 21(3), 2004
Introduction
Normal human gait is expected to be symmetrical and this statement was
supported by some authors who did not find any difference in ground reaction
forces between two legs during walking [3]. However there are some authors that
presented asymmetries in gait parameters [8]. These different opinions are mostly
related to the definition of gait symmetry and possibly to the variables selected to
assess it. But today many authors adhere to the school of thought that human gait is
not symmetrical and left and right legs have different functions [8,16]. On the other
hand it has been shown that there are some characteristic gait parameters for the
different patient groups and some sportsmen [2,5,7]. In the different branches of
sports, muscle groups are affected in different ways due to external results of
physically demanding exercises which ultimately lead to the shapenning of the
body.
Gait asymmetries may be used as a measurement criterion to differentiate
between normal and pathological gaits [4]. Furthermore they may be used during
the assessment of progress in the rehabilitation of the patients and sportsmen and
serve as an objective evaluation tool in the diagnosis of lower limb, pelvis and back
injuries [17].
The purpose of this study is to investigate and compare the gait asymmetries
between the normal population and football players by using insole mats. Insole
mats measure Ground Reaction Forces in Newton (N) and they were used
exclusively because they are objective and quantitative and also they have an
advantage according to the force platforms that it is possible to evaluate more than
one step for both sides in the same trial [6,14].
Material and Methods
Asymmetries in gait parameters were quantified using a Symmetry Index (SI)
that was proposed by Robinson et al.1987 [13].
SI 
1/ 2
XR  XL
100%
(X R  X L )
-
-
-
-
-
Where XR is a gait variable recorded for the right leg and XL is the corresponding
variable for the left leg. This index may be used for motion analysing systems,
Comparison of asymmetries
243
Biol.Sport 21(3), 2004
-
-
-
-
-
pressure distrubution devices, etc. A value of zero for SI indicates that there is no
difference between the variables XR and XL. In effect, this means that there exists a
perfect gait symmetry for the particular variable. A positive SI value indicates the
fact that XR>XL, the reverse is the case for a negative value.
Thirty-three males (mean age 20.6+/-1.1 yr; mean height 179.3+/-5.2 cm; mean
mass 71.2+/-6.8 kg) and 31 football players (mean age 22.1+/-1.8 yr; mean height
179.2+/-4.7 cm; mean mass 72.5+/-5.3 kg; active sport age 9.5+/-2.5 yr) were
informed about the procedure and accepted to participate in this study. Local ethics
comittee approved the procedure as well. They were all right-handed. Handedness
was checked by the “Edinburg Handedness Inventory” which separated absolute,
preferential or ambidextrous right or left handedness [11]. The first group
represented the normal human gait that had no history of neuromuscular or
musculoskeletal problems that may otherwise affect their gait parameters. In the
football players group this condition was checked by a sports physician to
eliminate acute phase problems.
Subjects wore boot-like designed flat shoes of Zebris© with insole-mats inserted
in it and were required to walk at natural speed along an 8 m-long footpath. After a
few trials of familiarization, the ground reaction forces were recorded from both
sides by Zebris 3D Motion Analysing System© [12]. This system has insole mats
connected to an analog digital convertor by a cable adaptor. Sampling rate was 60
Hz. Data converter was connected to a PC to enable the time versus the force
graphics to be seen while the subject was walking. To ensure standard data,
walking speed did not exceed or fall below 1.5 m/s. Therefore if the walktime was
less or more than 5.35+/-5% s, that trial was cancelled. It is possible to record 8
steps for adults on a footpath. Data from heel (1), middle foot (2), forefoot lateral
side (3), and forefoot medial side (4) were recorded separately in the same steps.
The heel was from 0% to 30% and midfoot from 30% to 60% to foot length. The
forefoot was from 60% to 100% and this part was divided equally from the middle
between the lateral and medial forefoot. Each of these areas are represented with a
time versus force graphic in the report paper. It is also possible to convert the data
as a textfile to process in a worksheet program.
To enable the deduction of detailed and accurate results, data were converted to
a textfile that avails the possibility of seeing the force values for every 3x10-3
seconds. Then for each variable, graphics were drawn and checked for gross errors.
The peak forces (Fmax1, Fmax2, Fmax3, Fmax4) and their corresponding times
from initial contact to each peak forces (Tmax1, Tmax2, Tmax3, Tmax4) were
recorded for every three steps and taking the four areas of the sole for both legs
into consideration as a unit of miliseconds (ms).The first and the last steps were not
244
B.S.Cigali et al.
Biol.Sport 21(3), 2004
assessed to eliminate any acceleration or deacceleration.To permit comparison
between the two groups, peak force values were normalized with the use of each
subject’s body weight [15,16]. Thus, the unit of measurement for the normalized
peak force was newtons per kilogram. The mean values were taken for all gait
variables and used for calculating XL and XR. Swing and Stance Phase Time
(phase%), Double Support Time (s) were also recorded.
SI values were calculated for the variables of 1. Swing and Stance Phase Time
2. Double Support Time 3. Fmax values 4. Tmax values. The Student’s t-test was
used to compare quantitative variables between right and left sides of each group
and between the two groups.
Results
Table 1
Mean values, mean symmetry indices (mean SI) and their standard deviations (SD)
for the selected gait variables of football players
Tswing (phase%)
Tstance (phase%)
DST(s)
Fmax1a
Fmax2a
Fmax3a
Fmax4a
Tmax1(ms)
Tmax2(ms)
Tmax3(ms)
Tmax4(ms)
Mean L
SD
38.22
61.78
0.14
5.3*
1.6
3.2*
2.7
157
290
598
636*
2.70
2.70
0.04
1.1
0.8
0.8
0.9
62.4
33.3
69.1
69.5
Mean
SD
R
39.38
2.19
60.63
2.19
0.13
0.04
4.2
0.91
1.3
0.52
2.9*
0.80
2.0
0.81
171
98.3
290
42.7
625
107.5
653
105. 1
P value Mean SI
(%)
0.062 -13.35
0.061 10.37
0.351
2.11
0.000 -24.62
0.023 -14.47
0.038 -16.20
0.004 -28.36
0.574 -7.71
0.810 -0.35
0.405 14.88
0.609
5.56
SD
(%)
8.45
11.90
8.93
15.12
16.98
23.09
6.78
20.50
16.56
14.08
15.54
a
Newtons Per kilogram
P values indicated in bold ithalics are mentioned as the significant differences
between the right and left sides of the same group while values that are bold with *
indicate the significant different values between both groups (P<0.05)
-
-
-
-
-
Analysis by groups: Table 1 shows mean values and symmetry index values for
temporal and kinetic gait variables in the case of the football players. P values are
245
Comparison of asymmetries
Biol.Sport 21(3), 2004
indicated in bold ithalics if there is a significant difference between the related
variables. For the Tswing, Tstance and DST, there were no significant differences
between the left and right side in the football players group. But kinetic variables
of left side (Fmax1, Fmax2, Fmax3, Fmax4) were found to be significantly higher
than the right side (P<0.05) while in the result of the Tmax values there were no
statistical differences.
Table 2
Mean values, mean symmetry indices (mean SI) and their standard deviations (SD)
for the selected gait variables of the control group
Tswing (phase%)
Tstance (phase%)
TDST(s)
Fmax1a
Fmax2a
Fmax3a
Fmax4a
Tmax1(ms)
Tmax2(ms)
Tmax3(ms)
Tmax4(ms)
Mean L
SD
Mean R
SD
39
61
0.13
5.86
1.39
2.73
3.05
152
275
564
602
2.70
2.70
0.03
1.18
0.64
0.98
0.98
42.7
111.7
50.5
45.3
39.66
60.34
0.12
4.57
1.22
2.34
2.47
142
271
564
609
2.29
2.29
0.02
0.94
0.41
0.69
0.74
60.0
96.4
53.4
53.6
P value Mean SI
(%)
0.290 -7.53
0.290
5.87
0.282
1.14
0.000 -39.97
0.171 -7.69
0.051 -15.45
0.003 -24.15
0.374 -26.30
0.832 -2.66
0.967 -0.45
0.434
6.01
SD
(%)
6.76
9.89
6.41
7.87
20.57
28.12
22.16
11.31
14.40
13.40
12.00
a
Newtons Per kilogram
P values indicated in bold ithalics show the significant differences between the
right and left sides
-
-
-
-
-
Data for the nonathletic control group were represented in Table 2. The left
Fmax1 and Fmax4 were statistically higher than the right side. No difference was
found for the other variables.
Analysis between the two groups: For the left side Fmax1, Fmax3, Tmax4 and
for the right side Fmax3 were found to be significantly different. These variables
are indicated in bold characters and * in Table 1. While Fmax1 is less for the
football players group, the values of the Fmax3 and Tmax4 are higher than the
control group.
246
B.S.Cigali et al.
Biol.Sport 21(3), 2004
Discussion
-
-
-
-
-
There are many studies to check whether the playing of a particular sport has
any influence on gait pattern [7]. Bessou et al. [1] found no differences in gait
between the right and left sides in healthy males and females of any age group.
Murray et al. [10] showed that walking was a highly productive activity. However,
playing football does not induce very different temporal gait parameters, some
differences of ground reaction forces were still deduced.
According to our results there is no exact symmetry for the football players and
the control group. As it is seen from the Tables 1 and 2 of the temporal parameters:
Tswing, Tstance and DST are more symmetric. The right side DST was found to be
longer in preferentially unilateral sports just because of the usage of left side as
impulse side when a sportsman needs to take a shot, dribble or take a jump [7]. But
in our study there is found to be no significant asymmetry for DST. The mean SI
values for these three parameters of two groups are close to zero and there is no
significant difference between the right and left side (P>0.05). This result is same
for the Tmax values of the four foot areas except Tmax4 for the left side of the
football players and the SI values are close to zero but there are no exact
symmetries. These results suggest that there is a close inter-individual
reproducibility of the human gait temporal parameters if the walking speed does
not vary.
For football players the Fmax values for the left side are higher and according
to the SI values for the Fmax1, Fmax2, Fmax3, Fmax4, there is an asymmetry
between the right and left legs (P<0.05). This difference can be explained by the
strength of the shock absorbent muscles of the right leg which work like brakes
when the leg touches on the floor. A right handed football player usually uses his
right foot to shoot and dribble, and his left foot to support the body before a jump
or a shot. The muscular development for football players has been found to be
asymmetric since the dominant right foot was found to be stronger than the left one
when measured by isokinetic dynamometers [9].
On the other hand only Fmax1 and Fmax4 of the left side of non-athletic control
group are statistically higher. This means control group hit their left heel and
medial forefoot strongly or in other words shock absorbent muscles of the nondominant side are weaker and they prefer to propel their bodies with the medial
forefoot for the non-dominant side. In the control group there is no difference
between the right and left sides for Fmax2 and Fmax3. This infers the absence of
any functional differences between the strength of the muscles which supports the
medial and lateral longitudinal foot arcs.
Comparison of asymmetries
247
Biol.Sport 21(3), 2004
When the two groups are compared in terms of their heel strikes, left Fmax1 is
less for the football players. The result is expected owing to the stronger shock
absorbent muscles of the football players. In the same token this also infers the fact
that the heel strikes of the football players group are softer than the control group.
There are significant differences for the Fmax3 values for both sides, but the mean
values are higher for the football players group, because this phase corresponds to
the terminal stance phase. During this phase the muscles are contracted to push
forward the body.
There is also a delay for Tmax4 for the non-dominant side of football players
because this group seems to touch their lateral forefeet quicker than the medial
side. Thus, this delay can be expected for the non-dominant side.
In conclusion football players can use their brake and propelling impulses better
because of their stronger muscular structures and they have an asymmetry in
ground reaction forces. The continuous playing of football over a period of years
appears to have resulted in some differences in terms of the ground reaction forces
between the right and left sides. It comes apparent the need for longitudinal studies
to be planned for both groups under evaluation. This in turn may reveal the
asymmetries and aid the organization of training programs.
Nowadays there exits scores of gait laboratories and study groups that have
their own methods. They utilize different systems for their gait evalutaion. SI
values reported by some authors fell between (6.3 to -6.3)% in the case of the
normal human gait [16}. In our study, we evaluated the maximum force values
independently for the four areas of the foot. In this respect we differ from other
researchers and the SI values range between (-7.69 and -39.97) % for the normal
human gait and (-14.47 to –28.36) % for the football players. As a result, the area is
inversely proportional to the asymmetry. So in most instances, the comparison
between the different laboratories is difficult when utilizing the direct values but
this symmetry index might come helpful when it comes to the evaluation and
comparison of various experimental inferences.
References
-
-
-
-
-
1. Bessou P., P.Dupui, R.Montoya, B.Pages (1988) Simultaneous recording of
longitudinal displacements of both feet during human walking. J.Physiol. (Paris) 83:
102-110
2. Cairns M.A., R.G.Burdett, J.C.Pisciotta, R.S.Simon (1986) A biomechanical
analysis of a race walking gait. Med.Sci.Sports Exerc. 18:446-453
3. Herzog W., B.M.Nigg, L.J.Read, E.Olsson (1989) Asymmetries in ground reaction
forces patterns in normal human gait. Med.Sci.Sports Exerc. 21:110-114
248
B.S.Cigali et al.
Biol.Sport 21(3), 2004
4. Jensen E.C., D.Vittas, S.Hellberg, J.Hansen (1982) Normal gait of young and old
men and woman. Ground reaction Force measurement on a treadmill. Acta Ortohopaed.
Scand. 53:193-196
5. Kelly V.E., M.J.Mueller, D.R.Sinacore (2000) Timing of peak plantar pressure
during the stance phase of walking. J.Am.Podiatr.Med.Assoc. 90:18-23
6. Kernozek T.W., K.A.Zimmer (2000) Reliability and running speed effects of inshoe loading measurements during slow treadmill running. Foot Ankle Int. 21:749-752
7. Leroy D., D.Polin, C.Tourny-Chollet, J.Weber (2000) Spatial and temporal gait
variable differences between basketball, swimming and soccer players. Int.J.Sports Med.
21:158-162
8. Maupas E., J.Paysant, N.Martinet, J.Andre (1999) Asymmetric leg activity in
healthy subjects during walking, detected by electrogoniometry. Clin.Biomech. 14:403-411
9. McLean B.D., D.M.Tumilty (1993) Left-right asymmetry in two types of soccer
kick. Br.J.Sports Med. 27:260-262
10. Murray M.P., R.C.Kory, B.H.Clarkson, S.B.Sepic (1996) Comparison of free and
fast speed walking patterns of normal men. Am.J.Phys.Med. 45:8-24
11. Oldfield R.C. (1971) The assesment and analysis of handedness. The Edinburg
inventory. Neuropsychol. 9:97-113
12. Rigas C. (1984) Spatial parameters of gait related to the position of the foot on the
ground. Prosthetics Orthotics Int. 8:130-134
13. Robinson R.O., W.Herzog, B.M.Nigg (1987) Use of force platform variables to
quantify the effects of chiropractic manipulation on gait symmetry. J.Manipulative Physiol.
Ther. 10:172-176
14. Sanfilippo P.B., R.M.Stess, K.M.Moss (1992) Dynamic plantar pressure analysis.
Comparing common insole materials. J.Am.Podiatr.Med.Assoc. 82:507-513
15. Simpson K.J., P.Jiang (1999) Foot landing position during gait influences ground
reaction forces. Clin.Biomech. 14:396-402
16. Van Zant R.S., T.G.McPoil, M.W.Cornwall (2001) Symmetry of plantar presures
and vertical forces in healthy subjects during walking. J.Am.Podiatr.Med.Assoc. 91:
337-342
17. Widhe T., L.Berggren (1994) Gait analysis and dynamic foot pressure in the
assessment of treated clubfoot. Foot Ankle Int. 15:186-190
-
-
-
-
-
Accepted for publication 1.10.2003