Walking speed, unilateral leg loading, and step symmetry in young adults

Gait & Posture 35 (2012) 66–69

Contents lists available at ScienceDirect

Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost

Walking speed, unilateral leg loading, and step symmetry in young adults Einat Kodesh *, Michal Kafri, Gali Dar, Ruth Dickstein Department of Physical Therapy, Faculty of Social Welfare and Health Sciences, University of Haifa, Mount Carmel, Haifa 31905, Israel

A R T I C L E I N F O

A B S T R A C T

Article history: Received 12 August 2010 Received in revised form 28 June 2011 Accepted 10 August 2011

The purpose of this study was to examine the effects of gait speed and unilateral lower limb loading on step time and step length symmetry in healthy adults. Spatiotemporal gait data were collected from 22 healthy subjects (11 men, 11 women), using the GaitRite walkway, under four randomly sequenced test conditions: self-selected speed (SS), fast speed (F), self-selected speed with the right leg loaded (LSS), and the fastest attainable speed with the right leg loaded (LF). The symmetry index, calculated with the formula [((R L)/0.5  (R + L))  100], was used to quantify step time and step length symmetry. It was found that over-ground gait speed had no significant effects on the symmetry of step time or step length. Unilateral lower limb loading significantly increased step time asymmetry, with longer step time for the loaded leg. Step symmetry was further compromised and became more asymmetrical when, in addition to unilateral leg loading, subjects maximized their gait speed. This effect of fast speed with unilateral leg loading was particularly prominent in relation to step length, with its shortening in the unloaded leg and lengthening in the loaded leg. These observations in healthy subjects may serve as a reference for the assessment of gait symmetry in patients with unilateral lower limb pathologies. ß 2011 Elsevier B.V. All rights reserved.

Keywords: Gait Walking Symmetry Unilateral load

1. Introduction In normal human locomotion, gait attributes of kinetic, kinematic, and spatiotemporal variables are assumed to be symmetrical [1,2] as manifested in matching kinematic and spatiotemporal characteristics of the left and right steps [3,4]. Slight asymmetry in one or several variables is considered within normal limits and may reflect functional differences in the contribution of each limb to propulsion and control during walking [2,5,6]. Conversely, gait asymmetry due to a temporary interference or a disorder of one lower limb is considered atypical and frequently indicates pathology. The relationships between corresponding gait variables measured for each limb are often expressed as indices of symmetry, providing numerical values of the degree of symmetry of the measured variables [6–8]. One factor that is known to affect gait variables is speed, and it is well established that walking should be tested under a range of speeds, rather than just self-selected speed [9,10]. However, most reports on over-ground gait symmetry in healthy individuals are limited to self-selected speed (e.g. [1,11,12]). An exception was Goble et al., who studied the symmetry of peak ground reaction forces and time to peak in twenty healthy adults during self-selected speed as well as during speeds that were 10% slower and 10% higher than the self-selected

* Corresponding author. Tel.: +972 4 8249065; fax: +972 4 8288140. E-mail addresses: [email protected], [email protected] (E. Kodesh). 0966-6362/$ – see front matter ß 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2011.08.008

speed. They found a high degree of inter-limb (between limbs) symmetry across the three speeds and recommended further research on the effect that greater walking speeds than those used in their study have on the symmetry of gait [13]. The symmetry of ground reaction force impulses was further tested in young healthy adults during slow, preferred, and fast walking. Overall, the findings pointed to symmetrical impulses, except for a 7% increase in the propulsive impulse of the dominant lower limb under faster speed conditions [14]. Notably, despite the fact that spatiotemporal step characteristics are outcome measures [15] reflecting overt gait characteristics, their inter-limb symmetry was not examined in these studies. Thus, the effect of different over-ground speeds, especially that of fast speed, on the step symmetry in healthy subjects is not well established. One method of augmenting gait asymmetry in healthy subjects is by loading one lower extremity [8,16]. A cast or other external attachment to one lower limb might have a similar effect; hence, in patients with unilateral limb pathology, the impact of an external load and the underlying pathology could be additive and vary across gait speeds. The effect of asymmetry induced by unilateral lower limb loading in healthy subjects has been examined in several treadmill studies. Haddad and colleagues studied intra- and inter-limb adaptations of spatiotemporal variables during self-selected speed. When asymmetrical walking was induced by increasing unilateral limb loading with six different loads, it was found that such unilateral distal loading mainly interfered with inter-limb, but not intra-limb, coordination [16]. In another study, Smith and Martin

E. Kodesh et al. / Gait & Posture 35 (2012) 66–69

used a speed of 1.57 m/s, which is slightly higher than the mean preferred walking speed of young adults, to study adaptation to a distally applied unilateral lower limb load. Overall, their results suggested that changes in temporal walking symmetry due to asymmetrical distal lower extremity loading are immediate and complete within 5 min of exposure to the load [8]. Notably, neither the load nor the speed was manipulated in this study. There are advantages to monitoring asymmetry via treadmill gait [17], and similarities between treadmill and over-ground walking have been demonstrated. However, the validity of generalizing findings from the treadmill to over-ground gait is not unequivocal [18–21], and any valid inference to over-ground gait should be supported by over-ground gait studies. In sum, the question of the effects of gait speed and unilateral lower limb loading on over-ground spatiotemporal gait symmetry in healthy subjects is not yet resolved. Therefore, the purpose of the current study was to examine gait speed (self-selected versus fast) and unilateral leg loading in order to determine the separate and the combined effects of these two variables on step time and step length symmetry in healthy individuals. 2. Methods 2.1. Setting and subjects Participants were 22 university students (11 men and 11 women), aged 27.2 (3.1) years, with no lower limb abnormalities or pathologies. Inclusion criteria also included right hand and leg dominance, as determined by self-reporting on manual dexterity and by the leg used for kicking a ball, which is highly correlated with lower limb dominance [22]. Exclusion criteria were cardiovascular or systemic ailments and leg length discrepancy exceeding 1 cm. The study was approved by the local institutional review board (IRB), and all participants signed informed consent forms. 2.2. Protocol Subjects were tested during the morning in a dedicated lab. Testing included four walking conditions with two speeds (self-selected and fast) and two modes of loading (‘‘no loading’’, which connoted unloaded, and ‘‘right leg loading’’). It is important to note that the term ‘‘self-selected’’ speed denotes the preferred comfortable gait speed of the individual, though in the current study the fast pace was also selected by the subject as the fastest speed attainable under the test conditions. For the loaded condition, a mass of 3 kg was fixed to the anterior distal part of the shank just above the lateral and medial malleoli and secured by velcro straps, similar to the manner applied by Smith and Martin [8]. A familiarization period of 20 min preceded data collection. During that period, each subject walked freely in the laboratory, initially at a self-selected speed, then at a fast speed with the load, and finally at a fast speed without the load. The familiarization period was followed by a rest period of 5 min. The GaitRite walkway, a reliable apparatus for measuring spatiotemporal gait data [22,23], was used for testing. The test included four conditions in random order: (1) self-selected speed (SS); (2) fast speed (F) (‘‘traverse the track as fast as you can’’); (3) self-selected speed with the attached mass (LSS); and (4) fast speed with the attached mass (LF). Four repetitions were performed for each condition. Each walking trial started at least 3 m before the measuring surface and terminated at least 2 m past it. In the two ‘‘fast’’ conditions, the examiner encouraged the subjects to walk as fast as possible. 2.3. Data handling and analysis The data from the last three trials collected via the GaitRite system were saved in Excel for analysis. Following inspection and descriptive analysis, the data were averaged for each participant. A symmetry index (SI) for step time and length was calculated using the formula: [((R L)/0.5  (R + L))  100] [2,8]. The SI was then used to create two indices: (1) the ‘‘directional’’ symmetry index, which was biased towards positive values when the values recorded from the right (loaded) lower limb were larger than those recorded from the left (unloaded) lower limb, and vice versa for the left limb; and (2) the ‘‘non-directional’’ or ‘‘absolute’’ symmetry index, which was the absolute value of the directional SI, calculated in order to prevent masking the true symmetry value (because positive and negative values of different subjects would cancel each other out, minimizing discrepancies). Descriptive statistics were used for the variables of interest. The effects of speed and right limb loading on the directional and absolute symmetry indices of step time and step length were studied using two-way ANOVA for repeated measures, with p values of 0.05 or lower considered significant. Further analysis of the results was done with Bonferonni’s post hoc testing and with paired t-tests. SPSS 18 software was used for analysis.

67

3. Results 3.1. Spatiotemporal characteristics The means and SDs of the spatiotemporal variables of interest are presented in Table 1. Results showed that the differences between the speed values were significant (F = 211, p  0.0001); post hoc Bonferonni testing indicated significantly higher speeds during each of the two fast speeds as compared to the two self-selected speeds under both loaded and unloaded conditions (p < 0.001 for all comparisons). No significant differences were noted between comparable speeds with and without loading. Similarly, the cadence values during the SS speed condition and during the LSS condition were not significantly different from each other. Likewise, the cadence values in the F and LF conditions were also similar. In each of the self-selected speed conditions (SS and LSS), the cadence was significantly lower than that associated with the fast speed conditions (F and LF) (p < 0.001 for all comparisons). The comparisons between the corresponding values for stride length yielded similar differences between the four conditions (p < 0.0001 for significant comparisons, see Table 1). As expected, the data in Table 1 show that with increasing gait speed, step time was shortened and step length increased. 3.2. Step time symmetry Results showed that the effect of speed on the directional symmetry index of step time was not significant, whereas the effect of load was significant (F = 13.3, p < 0.002). Without loading, the index was biased towards longer step time of the left extremity relative to the right extremity during both selfselected and fast speeds (note the negative values in Fig. 1). With loading, the index turned positive, indicating elongation of step time of the right loaded leg relative to the left unloaded leg. The interaction between speed and load was not significant, indicating a comparable effect of unilateral loading on step time symmetry during both self-selected and fast speeds. Further analysis with paired t-tests confirmed significant differences between the loaded and the unloaded conditions (p < 0.01 for all comparisons, see Fig. 1). The effect of speed on the absolute symmetry index of step time was also not significant, whereas the effect of load was associated with a significant increase in step time asymmetry (F = 6.04, p < 0.024). Further analysis with paired t-tests confirmed significant differences between the loaded and the unloaded conditions (p < 0.05, see Fig. 2). 3.3. Step length symmetry Results showed that the effect of speed on the directional symmetry index of step length was not significant. However, the effect of right leg loading was found to be significant (F = 9.5, Table 1 Spatiotemporal results with means and SDs in four walking conditions: selfselected (SS) speed, fast (F) speed, self-selected speed with the right leg loaded (LSS), and fast speed with the right leg loaded (LF). Condition

SS

F

LSS

LF

Speed cm/s Cadence step/min Stride length, cm Step time Rt, s Step time Lt, s Step length Rt, cm Step length Lt, cm

145.7 (17.7) 112 (8) 156.1(15.4) 0.53 (0.05) 0.54 (0.04) 77.94 (7.30) 77.42 (7.55)

268.1(33.2) 168 (18.1) 194.2 (16.2) 0.36 (0.04) 0.36 (0.05) 95.19 (8.23) 96.54 (8.37)

139.8 (19.8) 109 (7.9) 153.9 (17.5) 0.56 (0.05) 0.54 (0.04) 76.98 (7.99) 76.63 (8.99)

262.8 (34.7) 166 (19) 190.5 (18.4) 0.37 (0.05) 0.35 (0.04) 96.37 (9.15) 92.71 (9.03)

68

E. Kodesh et al. / Gait & Posture 35 (2012) 66–69

*

10

*

* *

SI

SI

*

4

*

5

*

6

2 0

0

-2

-5

Fig. 1. Step time: directional SI (%) at self-selected speed (SS), fast speed (F), selfselected speed with the right leg loaded (LSS), and fast speed with the right leg loaded (LF). Values are mean  SE (bars indicate significant differences between conditions).

p < 0.006), with a significant interaction between speed and loading (F = 13.5, p < 0.001), indicating that the effect of loading on step length symmetry was primarily manifested in the LF condition. Further analysis with paired t-tests substantiated significant differences between the LF condition and the SS, F, and LSS conditions (p < 0.025, 0.0001, and 0.042, respectively). The data presented in Fig. 3 depict the exceptional high asymmetry that was created by the unilateral loading during fast walking (LF condition). The effect of speed on the absolute symmetry index of step length was also not significant, whereas loading was found to have a significant effect (F = 9.5 p < 0.006), with a significant interaction between the effects of load and speed (F = 5.5 p < 0.03) (see Fig. 4). Further analysis with paired t-tests confirmed significant differences between the LF condition and the SS, F, and LSS conditions (p < 0.009, 0.002, and 0.04, respectively). Thus, the findings for the absolute symmetry index concurred with those of the directional index, indicating increased step length asymmetry with right lower extremity loading during fast walking. 4. Discussion The findings indicate that during indoor ambulation, fast walking speed had no significant effect on the symmetry of step time and length. Thus, it appears that in healthy young adults, the

*

15

*

LF

F

LS S

F L

LS

S

F

SS

-10

SS

-4

Fig. 3. Step length: directional SI (%) at self-selected speed (SS), fast speed (F), selfselected speed with the right leg loaded (LSS), and fast speed with the right leg loaded (LF). Values are mean  SE (bars indicate significant differences between conditions).

automatically chosen magnitude of fast self-selected speed is such that right–left step symmetry, an innate basic characteristic of human gait [3,4,24], is preserved. Speed increase is associated with shortening of stride time and/ or with lengthening of stride length. Although the findings indicated that during fast speed these changes were comparable for the right and left steps, they pointed to differences between step time and length with unilateral leg loading. As loading affected these parameters in both legs, we regard the changes in step time and length of the right loaded leg as primary and those in the left contralateral leg as compensatory. The significant changes in the directional symmetry index of step time indicate a relative elongation of step time of the right loaded leg and/or a relative decrease in step time of the left unloaded leg. Based on the data (Table 1), the lengthening of step time of the loaded leg was more ubiquitous than the shortening of step time of the unloaded leg. As the added mass increased the moment of inertia of the loaded leg during swing [25], slowed deceleration to initial contact explains the lengthening of step time. The compensatory shortening of step time on the unloaded side probably stemmed from an attempt to reduce stride time and maintain speed, especially during the LF condition. More pronounced is the observation that the combination of unilateral leg loading and fast gait (LF condition) significantly and dramatically affected the symmetry of step length with the

*

SI

10

5

0

SS

F

S LS

LF

Fig. 2. Step time: absolute SI (%) at self-selected speed (SS), fast speed (F), selfselected speed with the right leg loaded (LSS), and fast speed with the right leg loaded (LF). Values are mean  SE (bars indicate significant differences between conditions).

Fig. 4. Step length: absolute SI (%) at self-selected speed (SS), fast speed (F), selfselected speed with the right leg loaded (LSS), and fast speed with the right leg loaded (LF). Values are mean  SE (bars indicate significant differences between conditions).

E. Kodesh et al. / Gait & Posture 35 (2012) 66–69

enhancement of step length of the loaded extremity and with the shortening of step length of the unloaded extremity. The enhancement of step length of the loaded leg is explained by higher than normal upward and forward momentum at the initial swing, as well as by delayed and slowed deceleration at terminal swing. As swing duration largely determines step length [26], the observed increase in step length is not surprising and conforms with prior observations on the effect of lower leg loading [8,25,27,28]. Noteworthy, however, is the demonstration that the major change between the LSS and the LF conditions was the compensatory decrease in step length of the unloaded leg (see Table 1). Since AP balance during gait is largely dependent on step length, with shorter steps associated with better balance [29], it appears that the shortening of step length of the unloaded leg counterbalanced the step lengthening of the loaded leg, thus maintaining optimal stride length during the LF condition. In other words, the increase in step length asymmetry was essential for maintaining optimal stride length to generate the fastest possible walking speed. Overall, the bilateral changes in step time and length symmetry could emanate from the intention to carry out the demanded tasks with stride time and length (i.e., of both extremities) that minimize metabolic energy consumption [25,30]. These changes also reflect the adaptability of stride time and length to the requirements of the task (in the extreme condition, to the imposed unilateral load with the demand to walk as fast as possible). The findings lend support to prior evidence that intentional control of stride length is a fundamental basis for the adaptation of stride to environmental constraints and/or to task requirements [31]. 5. Conclusions The current demonstration of spatiotemporal asymmetry associated with unilateral leg loading supplements prior reports of increase in kinetic and kinematic asymmetry by unilateral lower limb loading during both over-ground and treadmill walking. Since spatiotemporal step asymmetry in healthy adults is not related to age [32], we presume that these findings can be generalized to different age groups of healthy adults. These observations in healthy populations may serve as a reference for the assessment of gait symmetry in patients with unilateral lower limb pathologies. The current findings help to shed light on the nature of asymmetrical gait in patients, especially among those individuals who compromise symmetry for increasing gait speed. In order to elaborate on the relationship between walking speed and symmetry, patients should be tested using within-subject designs under natural over-ground walking conditions. Initial steps in this direction have already been taken, and the relevant questions should be further explored [32–34]. Conflict of interest There is no conflict of interest of any author of the paper. References [1] Lythgo N, Wilson C, Galea M. Basic gait and symmetry measures for primary school-aged children and young adults whilst walking barefoot and with shoes. Gait Posture 2009;30(November (4)):502–6. [2] Sadeghi H. Local or global asymmetry in gait of people without impairments. Gait Posture 2003;17(June (3)):197–204. [3] Butt S, Lebret J, Kiehn O. Organization of left–right coordination in the mammalian locomotor network. Brain Res Brain Res Rev 2002;40:107–17.

69

[4] Ivanenko YP, Poppele RE, Lacquaniti F. Distributed neural networks for controlling human locomotion: lessons from normal and SCI subjects. Brain Res Bull 2009;78(January (1)):13–21. [5] Imamura M, Imamura ST, Salomao O, Pereira CA, De Carvalho Jr AE, Neto RB. Pedobarometric evaluation of the normal adult male foot. Foot Ankle Int 2002;23(September (9)):804–10. [6] Sadeghi H, Allard P, Prince F, Labelle H. Symmetry and limb dominance in ablebodied gait: a review. Gait Posture 2000;12(September (1)):34–45. [7] Patterson KK, Gage WH, Brooks D, Black SE, McIlroy WE. Evaluation of gait symmetry after stroke: a comparison of current methods and recommendations for standardization. Gait Posture 2010;31(February (2)):241–6. [8] Smith JD, Martin PE. Walking patterns change rapidly following asymmetrical lower extremity loading. Hum Mov Sci 2007;26(June (3)):412–25. [9] Beaman CB, Peterson CL, Neptune RR, Kautz SA. Differences in self-selected and fastest-comfortable walking in post-stroke hemiparetic persons. Gait Posture 2010;31(March (3)):311–6. [10] Bloem B, Duysens J. Pererring preferred speed? Paper presented at XIX Conference of the International Society for Posture & Gait, Bologna; 2009. [11] Gundersen LA, Valle DR, Barr AE, Danoff JV, Stanhope SJ, Snyder-Mackler L. Bilateral analysis of the knee and ankle during gait: an examination of the relationship between lateral dominance and symmetry. Phys Ther 1989;69 (August (8)):640–50. [12] Herzog W, Nigg BM, Read LJ, Olsson E. Asymmetries in ground reaction force patterns in normal human gait. Med Sci Sports Exerc 1989;21(February (1)): 110–4. [13] Goble DJ, Marino GW, Potvin JR. The influence of horizontal velocity on interlimb symmetry in normal walking. Hum Mov Sci 2003;22(August (3)): 271–83. [14] Seeley MK, Umberger BR, Shapiro R. A test of the functional asymmetry hypothesis in walking. Gait Posture 2008;28(July (1)):24–8. [15] Winter DA. The biomechanics and motor control of human gait, 2nd ed., Waterloo: University of Waterloo; 1991. [16] Haddad JM, van Emmerik REA, Whittlesey SN, Hamill J. Adaptations in interlimb and intralimb coordination to asymmetrical loading in human walking. Gait Posture 2006;23(June (4)):429–34. [17] Draper ER. A treadmill-based system for measuring symmetry of gait. Med Eng Phys 2000;22(April (3)):215–22. [18] Lee SJ, Hidler J. Biomechanics of overground vs. treadmill walking in healthy individuals. J Appl Physiol 2008;104(March (3)):747–55. [19] Terrier P, Deriaz O. Kinematic variability, fractal dynamics and local dynamic stability of treadmill walking. J Neuroeng Rehabil 2011;24(February):8. [20] Wank V, Frick U, Schmidtbleicher D. Kinematics and electromyography of lower limb muscles in overground and treadmill running. Int J Sports Med 1998;19(October (7)):455–61. [21] Watt JR, Franz JR, Jackson K, Dicharry J, Riley PO, Kerrigan DC. A threedimensional kinematic and kinetic comparison of overground and treadmill walking in healthy elderly subjects. Clin Biomech 2010;25(June (5)):444–9. [22] Kalaycioglu C, Kara C, Atbasoglu C, Nalcaci E. Aspects of foot preference: differential relationships of skilled and unskilled foot movements with motor asymmetry. Laterality 2008;13(March (2)):124–42. [23] McDonough AL, Batavia M, Chen FC, Kwon S, Ziai J. The validity and reliability of the GAITRite system’s measurements: a preliminary evaluation. Arch Phys Med Rehabil 2001;82(March (3)):419–25. [24] Yakovenko S, McCrea DA, Stecina K, Prochazka A. Control of locomotor cycle durations. J Neurophysiol 2005;94(August (2)):1057–65. [25] Browning R, Modica J, Kram R, Goswami A. The effects of adding mass to the legs on the energetics and biomechanics of walking. Med Sci Sports Exerc 2007;39:515–25. [26] Whittle MW. Gait analysis: an introduction, 2nd ed., Oxford: ButterworthHeinemann; 1996. [27] Eke-Okoro S, Larsson L, Sandlund B. Simulation of paretic gait in normal subjects by loading the ankles. Scand J Rehabil Med 1985;17:147–50. [28] Eke-Okoro ST. Exploration of paretic gait by differential loading in normals. Clin Biomech 1999;14(February (2)):136–40. [29] Neuman D. Kinesiology of the muskuloskeletal system. St. Louis: Mosby; 2002. [30] Russell DM, Kalbach CR, Massimini CM, Martinez-Garza C. Leg asymmetries and coordination dynamics in walking. J Mot Behav 2010;42(May–June (3)):157–68. [31] Varraine E, Bonnard M, Pailhous J. Intentional on-line adaptation of stride length in human walking. Exp Brain Res 2000;130(January (2)):248–57. [32] Patterson KK, Parafianowicz I, Danells CJ, Closson V, Verrier MC, Staines WR, et al. Gait asymmetry in community ambulating stroke survivors. Arch Phys Med Rehabil 2008;89(February (2)):304–10. [33] Alexander LD, Black SE, Patterson KK, Gao F, Danells CJ, McIlroy WE. Association between gait asymmetry and brain lesion location in stroke patients. Stroke 2009;40(February (2)):537–44. [34] Khanna I, Roy A, Rodgers MM, Krebs HI, Macko RM, Forrester LW. Effects of unilateral robotic limb loading on gait characteristics in subjects with chronic stroke. J Neuroeng Rehabil 2010;7:23.