Neuroscience Letters 285 (2000) 150±154
www.elsevier.com/locate/neulet
Is human gait initiation program affected by a reduction of the postural basis? Annabelle Couillandre*, Yvon BrenieÁre, Bernard Maton Laboratoire de Physiologie du Mouvement, INSERM U483, Universite Paris-Sud, Bat 441, 91405 Orsay, France Received 14 January 2000; received in revised form 17 March 2000; accepted 17 March 2000
Abstract The aim of this study was to access the adaptability of the gait initiation program by imposing before and during gait a posture that partially prevents the backward shift of the center of foot pressure. Six healthy subjects performed normal gait in the control situation (CS) and gait in the absence of heel ground contact in the test situation (TS) on a force platform at three different speed conditions. It is shown that an increase in the duration of the anticipation phase in TS is necessary to create conditions for progression which allow the subjects to reach a gait velocity similar to the one obtained in CS at the end of the anticipatory movements and also at the end of the ®rst step. Modi®cations of the gait initiation program occur in order to ful®l the performance in terms of gait velocity. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Gait initiation; Locomotion; Postural basis; Center of foot pressure; Center of gravity; Velocity
During the anticipation phase of the gait initiation process which is the transient period between upright posture and steady state gait, the central nervous system has to create the dynamics necessary to induce the appropriate disequilibrium for reaching a given gait velocity. Both muscular and gravitational actions are necessary to create the postural and dynamic conditions for progression [17,19,20]. A forward and controlateral acceleration of the center of gravity (CG) and a backward and ipsilateral displacement of the center of foot pressure (CP) before the intentional movement are the earlier changes in posture and constitute anticipatory postural adjustments (APA) [1,7,18,21]. The duration and the amplitude of the APA in the sagittal plane are predictive of the future gait velocity, setting up the concept of a locomotor program [3,9,10]. Knowing that the duration of the initiation phase is invariant being dependent on individual biomechanical constants [2,6,12], faster gait requires a longer anticipation time and a shorter execution time. The velocity of CG at the end of the anticipation phase establishes a linear relation with the velocity of CG at the end of the initiation phase [4,14]. Without this initial
* Corresponding author. Tel.: 133-1-6915-5863; fax: 133-16915-5869. E-mail address:
[email protected] (A. Couillandre).
velocity at time of heel off, the subjects are not able to reach the gait velocity within one step. The CP, the barycenter of the vertical ground reaction forces, is a relevant element of the gait initiation process. During the anticipation phase, a linear covariation has been established between its maximum backward shift and the gait velocity reached at the end of the ®rst step [4,14,15]. Nevertheless, the backward shift of CP is limited because it can not go out of the support basis. In other words, the APA prior to a voluntary movement are dependent on the initial posture [4,8,16]. The progression velocity can be taken as an index of performance. This velocity has two different expressions [3,6]. The ®rst one, the instantaneous velocity of CG, results on the integration of all the propulsive forces acting on the body which are contained in the ground reaction. Its value is an expression of the peripheral effects of external and internal forces, the external forces being the gravity and the internal forces being the muscular actions. The second one, the progression velocity of CP, is calculated from the ®rst step length and frequency and may be considered to be a re¯ection of the displacement of the postural basis of support. During gait, these two expressions of gait velocity have to be controlled in order to be equal at the end of the ®rst step. The present work was designed to access the adaptability of gait initiation by analyzing how this process is modi®ed
0304-3940/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S0 30 4- 39 40 ( 00) 0 10 15- 6
A. Couillandre et al. / Neuroscience Letters 285 (2000) 150±154
when a body posture which does not allow a normal backward shift of CP is imposed before and during gait. How does the healthy subject compensate the restrictions linked to the initial body posture to perform the gait initiation program? In other words, is the healthy subject able to achieve a gait velocity similar to the one described in normal gait initiation when he walks with his heels slightly off the ground? A force platform AMTI LG6-4-1 was used, like in previous studies [4], to compute the accelerations of CG and to measure the shifts of CP from the ground reaction forces and moments. The anteroposterior velocity of CG was calculated from its anteroposterior acceleration. Signals from the force platform were stored on the disk of a PC microcomputer using a digitized frequency of 500 Hz. Six healthy subjects took part in the experimentation. Each subject stood upright, barefoot and motionless on the force platform. His feet's position on the force platform was noted in order to ®nd the position of CP. Two situations were examined. In the ®rst one which is the control situation (CS), gait initiation was performed normally from an erected spontaneous posture; in the second one which is the test situation (TS), the subject was instructed to keep the heels slightly off the ground before and during gait. Each subject performed three series of seven gait movements in accordance with three different speed instructions: slow, normal and fast. The biomechanical parameters considered for this study were (Fig. 1): (i) the time (t0) of the onset of the ®rst mechanical phenomena; the time (tTO) of Toe Off of the stepping foot; the time (tV) of the ®rst step execution end, (ii) the average initial position of CP (xP0) at t0 (Table 1), (iii) the peak amplitude of the backward shift of CP (MinxP) during the anticipation phase, (iv) the anteroposterior velocity of CG at tTO (x 0 TO) and at tV (V), (v) the length (L) and frequency (f) of the ®rst step which de®ne the mean anteroposterior velocity of CP (v Lf ). Mean values and standard deviations of the biomechanical parameters of gait initiation were calculated for each speed condition. The in¯uence of the initial body posture on the different variables of gait initiation was computed using a Student t-test for the spontaneous speed condition for paired data. The effect of gait velocity on the considered parameters was studied by mean of a Fischer±Snedecor test. Linear correlations and coef®cients of correlation were calculated to test the degree of coherence between gait velocity and the other biomechanical parameters. Typical mechanical traces of the two experimental situations are shown on Fig. 1. The main difference between TS and CS concerned the anteroposterior acceleration of CG (x 00 G) and the displacement of CP (xP and yP). The mechanical traces in TS are characterized by a weak initial anteroposterior acceleration of CG at t0,which contrasts with the well established increase of the acceleration of CG at that same time in CS. Moreover in 34% of trials, while anteroposterior and lateral shifts of CP occur at t0 in CS, the
151
Fig. 1. Recordings of the biomechanical parameters during control situation (CS) and test situation (TS); one trial, one subject, normal speed for each situation. x 00 G, x 0 G, acceleration and velocity of CG along the anteroposterior axis; xP and yP, anteroposterior and lateral displacement of CP with 0 corresponding here to the initial position of CP before movement. t0, time of the onset of the ®rst mechanical phenomena; MinxP, peak amplitude of the backward shift of CP during the anticipation phase; tTO, time of Toe Off of the stepping foot; x 0 TO, anteroposterior velocity of CG at tTO; L, length of the ®rst step; T/2, half period of the ®rst step; tV, time of the ®rst step execution end; V, anteroposterior velocity of CG at tV. a, postural phase; b, anticipation phase; c, execution phase.
mechanical recordings in TS associated a weak anteroposterior displacement of CP with no lateral displacement at that particular time. In all trials related to TS, the subjects took more time to organize the intentional movement of Toe Off and consequently lengthened the duration of the anticipation and initiation phase when they stood on their forefeet. Whatever the speed condition, a signi®cant difference (assessed by a Student t-test and a Fischer±Snedecor test (P , 0:01)) was found between CS and TS for the average initial position of CP (xP0) before movement (F(1,35) 193.49), the peak amplitude of the backward shift of CP (MinxP) during the anticipation phase (F
1; 35 23), the duration of the anticipation and initiation phase (tTO and tV) (Table 1). For the duration of the anticipatory movements, F(1,35) was 39.6 and 14.85 for the duration of the initiation phase. In particular in the spontaneous speed condition, results showed that the CP before movement was located 80 mm more ahead in TS than in CS. This forward position of CP implied some important initial postural constraints for the whole body segments because the ground projection of CG
152
A. Couillandre et al. / Neuroscience Letters 285 (2000) 150±154
Table 1 Mean values and standard deviations of the parameters of the gait initiation process in control situation (CS) and in test situation (TS) a Speed instructions
Slow
Normal
Fast
CS
TS
Signi®cance P , 0.01
CS
TS
Signi®cance P , 0.01
CS
TS
Signi®cance P , 0.01
xP0 (mm) MinxP (mm) tTO (s) x 0 TO (m/s) L (m) f (Hz) v Lf (m/s) tV (s) V (m/s)
52 ^ 14 245 ^ 11 0.687 ^ 0.108 0.293 ^ 0.078 0.583 ^ 0.085 1.646 1 0.158 0.966 ^ .0.209 1.238 ^ 0.076 0.961 ^ 0.179
135 ^ 18 225 ^ 6 0.846 ^ 0.120 0.311 ^ 0.070 0.6 ^ 0.101 1.904 ^ 0.320 1.126 ^ 0.180 1.307 ^ 0.064 0.998 ^ 0.136
S S S NS NS NS NS S NS
57 ^ 14 266 ^ 8 0.62 ^ 0.080 0.386 ^ 0.085 0.676 ^ 0.082 1.874 ^ 0.203 1.27 ^ 0.233 1.109 ^ 0.049 1.233 ^ 0.162
137 ^ 18 237 ^ 9 0.836 ^ 0.106 0.395 ^ 0.079 0.669 ^ 0.065 2.103 ^ 0.243 1.404 ^ 0.186 1.254 ^ 0.086 1.228 ^ 0.129
S S S NS NS NS NS S NS
62 ^ 18 290 ^ 20 0.567 ^ 0.092 0.491 ^ 0.132 0.747 ^ 0.099 2.15 ^ 0.420 1.624 ^ 0.490 1.015 ^ 0.073 1.613 ^ 0.199
133 ^ 22 250 ^ 11 0.816 ^ 0.073 0.548 ^ 0.074 0.743 ^ 0.082 2.451 ^ 0.240 1.819 ^ 0.282 1.187 ^ 0.070 1.549 ^ 0.179
S S S NS NS NS NS S NS
a xP0, average initial position of CP before movement; MinxP, peak amplitude of the backward shift of CP during the anticipation phase; tTO, time of Toe Off of the stepping foot; x 0 TO, anteroposterior velocity of CG at tTO; L, length of the ®rst step; f, frequency of the ®rst step; v, mean anteroposterior velocity of CP; tV, time of the ®rst step execution end; V, anteroposterior velocity of CG at tV. S, signi®cant at P , 0:01; NS, non signi®cant at P , 0:01.
de®nitely has to coincide with the position of CP [1]. The amplitude of the backward shift of CP during the anticipation phase was less important in TS for the three speed conditions showing that the subjects have followed the postural orders given by the experimenter during the experimentation (Table 1). The initial posture characterized by an absence of heel ground contact ful®ls its speci®c goal because it partially prevents the backward displacement of CP. Whatever the speed instruction, the consequence of this imposed postural constraint principally appeared in the duration of the anticipatory movements and consequently in the duration of the initiation phase but not in the duration of the step execution phase. The anticipation phase was longer in TS, its increase consequently and signi®cantly lengthened the duration of the initiation phase (Table 1). A longer anticipation phase allowed the subjects to reach a velocity at time of Toe Off, x 0 TO, which was not signi®cantly different from TS to CS (Table 1) and which like in normal gait initiation covaried with the forthcoming gait velocity, V (Fig. 3a). The coef®cient of correlation was 0.721 in TS versus 0.885 in CS (Fig. 3a). Reaching this velocity at the end of the anticipation phase enabled the subjects to create convenient conditions for progression as soon as time of Toe Off and allowed them to develop a progression velocity, either characterized by the velocity of CP, v, or the instantaneous velocity of CG at the end of the ®rst step, V, not signi®cantly different between TS and CS (Table 1). The step execution phase was not modi®ed: results showed a conservation of the characteristic locomotor parameters. The ®rst step length, L, and frequency, f, showed some variations that were not signi®cantly different between CS and TS, for each speed condition (Table 1). These parameters were also correlated with the anteroposterior velocity of CG at tV. The step execution phase is a ballistic phase [14] which allows the determination of the velocity of CG at the end of the ®rst step from its value at the end of the APA. In both situations, the covariations established between
the peak amplitude of the backward shift of CP during the anticipation phase and the anteroposterior velocity of CG at the end of the initiation phase relate the initial postural basis to the ®nal performance of gait velocity. Coef®cients of correlations were, respectively, 0.845 in CS and 0. 617 in TS (Fig. 2). However, if a comparison between the slopes of the linear regressions was established, the covariations were partly different probably because the support basis was different. In TS, faster gait still required a larger backward shift of CP even if this shift was experimentally partially inhibited. In both situations, a direct proportionality between the mean anteroposterior velocity of CP, v, and the velocity of CG at the end of the ®rst step, V, was expressed since a linear regression existed between the two parameters (vCS 1:02V CS with r CS 0:860; vTS 1:156V TS with r TS 0:937) (Fig. 3b). A signi®cant difference (assessed by a Student t-test (P , 0:01)) was found between CS and TS as far as the slopes of the linear regressions were analyzed. Unlike in CS, the two different expressions of the progression velocity were not exactly equal at the end
Fig. 2. Relationship between the anteroposterior velocity of CG at the end of the ®rst step (V) and the peak amplitude of the backward shift of CP during the anticipation phase (MinxP) in control situation (CS) and in test situation (TS). r, correlation coef®cient.
A. Couillandre et al. / Neuroscience Letters 285 (2000) 150±154
Fig. 3. (a) Relationship between the anteroposterior velocity of CG at the end of the ®rst step (V) and the anteroposterior velocity of CG at time of Toe Off (x 0 TO). (b) Relationship between the anteroposterior velocity of CG at the end of the ®rst step (V) and the mean anteroposterior velocity of CP (v) in control situation (CS) and in test situation (TS). r, correlation coef®cient.
of the initiation phase in TS. The mean anteroposterior velocity of CP was always greater than the velocity of CG at the end of the ®rst step, which indicated that the subjects did not exactly adjust the progression of CG to the progression of CP at the end of the ®rst step. Previous studies have shown that the aim of the gait initiation process was to place the subjects in steady-state gait at the end of the ®rst step, in an invariant time, by means of APA [2]. These latter are essential because they contribute to create the convenient conditions for walking [3]. Lepers et al. have analyzed changes of the gait initiation program following a running exercise; they have observed that the invariances and characteristics of gait initiation can be modi®ed by a locomotor activity [15]. In the present study, we would like to analyze if and how the gait initiation program is changed by a reduced con®guration of the support basis before and during gait. Previous studies on normal gait initiation demonstrated that the generation of an anteroposterior acceleration of the center of mass depends on the CP backward shift [1,4,9,20]. During the anticipation phase, an appropriate gap between the position of CP and the projection onto the ¯oor of the center of mass has to be created in order to initiate the forward fall. This forward fall is a speci®c postural adjustment which anticipates the step execution. In TS, the subjects compensate for the reduction of the initial support basis by adopting a particular anticipation strategy. The increase in the duration of the anticipatory movements observed in the results leads to a longer voluntary forward fall of the subjects which allows them to reach already at time of Toe off a velocity of CG similar to the one observed in CS. Like in normal gait initiation [4,14], this velocity at the end of the APA predetermines the gait velocity, V,
153
which is not signi®cantly different to the one observed in CS even if gait is initiated from a posture that drastically limits the backward shift of CP but however does not diminish the capacity of action of the internal and external forces such as gravity. Nevertheless in TS, the anteroposterior velocity of CG and the mean anteroposterior velocity of CP at the end of the ®rst step are not equal. In TS, the velocity of CP is always greater than the velocity of CG. BrenieÁre et al. have shown that the steady-state gait is reached within one step in adults [2] but not in children [5]. Children reach the steady state velocity after two to four steps. In normal gait initiation, the initial forward fall is necessary to get an initial displacement of CG and is followed by the execution of the ®rst step which is programmed to adjust the progression of CP to the one of CG [6]. If the mean anteroposterior velocity of CP is a combined control of the length and frequency of the ®rst step, the results suggest that the subjects in TS de®nitely overestimate the displacement of the support basis. This result would imply that the subjects do not have the postural and dynamic capacities to adjust the progression of CG to the one of CP within one step because they probably do not manage the propulsives forces (muscular and gravitational) during the anticipation phase as precisely as in CS. This lack of estimation of the muscular and gravitational forces could come from the exteroceptive and proprioceptive disturbance imposed by the initial body posture. Indeed, among the exteroceptive and proprioceptive information, those arising from the plantar sole are known to play an important role in posture and gait control [13] and are at evidence altered by the heel-off initial posture. The mechanism responsible for the changes in the gait initiation process observed in the results could be similar to that suggested by Lepers et al. for the adaptation of the postural phase of gait initiation which occur after a running exercise [15]: a change in the body schema, in the sense of a central dynamic representation of the body in its environment [11]. This work was performed in Institut de Myologie, Groupe Hospitalier Pitie - SalpeÃtrieÁre, Paris, France, and was partly supported by funds from AFM. [1] Breniere, Y., Do, M.C. and Sanchez, J., A biomechanical study of the gait initiation process, J. Biophys. Med. Nucl., 5 (1981) 197±205. [2] Breniere, Y. and Do, M.C., When and how does steady state gait movement induced from upright posture begin? J. Biomech., 19 (1986) 1035±1040. [3] Breniere, Y., Do, M.C. and Bouisset, S., Are dynamic phenomena prior to stepping essential to walking? J. Mot. Behav., 19 (1987) 62±76. [4] Breniere, Y. and Do, M.C., Modi®cations posturales associeÂes au lever du talon dans l'initiation du pas de la marche normale, J. Biophys. Biomec., 11 (1987) 161±167. [5] Breniere, Y., Bril, B. and Fontaine, R., Analysis of the transition from upright stance to steady state locomotion in children with under 200 days of autonomous walking, J. Mot. Behav., 21 (1989) 20±37.
154
A. Couillandre et al. / Neuroscience Letters 285 (2000) 150±154
[6] Breniere, Y. and Do, M.C., Control of gait initiation, J. Mot. Behav., 23 (1991) 235±240. [7] Cook, T. and Cozzens, B., Human solutions for locomotion: the initiation of gait, In R.M., Herman, S. Grillner, P.S.G. Stein and D.G. Stuart (Eds.), Neural Control of Locomotion, Plenum Press, New York, 1976, pp. 65±76. [8] Cordo, P.J. and Nashner, L.M., Properties of postural adjustments associated with rapid arm movements, J. Neurophysiol., 47 (1982) 287±302. [9] Crenna, P. and Frigo, C., A motor programme for initiation of forward oriented movements in man, J. Physiol. (Lond.), 437 (1991) 635±653. [10] Dietrich, G., Breniere, Y. and Do, M.C., Organization of local anticipatory movements in single step initiation, Hum. Mov. Sci., 13 (1994) 195±210. [11] Gur®nkel, V.S., Levik, Yu.S., Popov, K.E., Smetanin, B.N. and Shlikov, Yu.V., Body scheme in the control of postural activity, In V.S., Gur®nkel, M.E. Ioffe, J. Massion and J.P. Roll (Eds.), Stance and Motion, Facts and Concepts, Plenum Press, New York, 1988, pp. 185±193. [12] Jian, Y., Winter, D.A., Ishac, M.G. and Gilchrist, L., Trajectory of the body COG and COP during initiation and termination of gait, Gait Posture, 1 (1993) 9±22. [13] Kavounoudias, A., Roll, R. and Roll, J.P., Speci®c wholebody shifts induced by frequency-modulated vibrations of human plantar soles, Neurosci. Lett., 266 (1999) 181±184.
[14] Lepers, R. and Breniere, Y., The role of anticipatory postural adjustments and gravity in gait initiation, Exp. Brain Res., 107 (1995) 118±124. [15] Lepers, R., Breniere, Y. and Maton, B., Changes to the gait initiation programme following a running exercise in human subjects, Neurosci. Lett., 260 (1999) 69±73. [16] Lipshits, M.I., Mauritz, K. and Popov, K.E., Quantitative analysis of anticipatory postural components of a complex voluntary movement, Fiziologiya Cheloveka, 7 (1981) 411± 419. [17] Mackinnon, C.D. and Winter, D.A., Control of whole body balance in the frontal plane during human walking, J. Biomech., 26 (1993) 633±644. [18] Mann, R.A., Hagy, J.L., White, V. and Liddell, D., The initiation of gait, J. Bone Joint Surg., 61 (1979) 232±239. [19] Prince, F., Winter, D.A., Stergiou, P. and Walt, S.E., Anticipatory control of upper body balance during human locomotion, Gait Posture, 2 (1994) 19±25. [20] Winter, D.A., Mackinnon, C.D., Ruder, G.K. and Wieman, C., An integrated EMG/biomechanical model of upper balance and posture during human gait, Prog. Brain Res., 97 (1993) 359±367. [21] Yamashita, T. and Katoh, R., Moving pattern of point of application of vertical resultant force during level walking, J. Biomech., 9 (1976) 93±99.