- Email: [email protected]
In-shoe plantar pressure distribution and lower extremity muscle activity patterns of backward compared to forward running on a treadmill
Gait & Posture 46 (2016) 135–141
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Full length article
In-shoe plantar pressure distribution and lower extremity muscle activity patterns of backward compared to forward running on a treadmill Thorsten Sterzing a,c, Clivia Frommhold b,c, Dieter Rosenbaum d,* a
Li Ning Sports Science Research Centre, Beijing, China Department Medical Affairs, Bauerfeind AG, Zeulenroda, Germany Department of Human Locomotion, Chemnitz University of Technology, Chemnitz, Germany d Movement Analysis Lab, Institute for Experimental Musculoskeletal Medicine, University Hospital Mu¨nster, Mu¨nster, Germany b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 September 2015 Received in revised form 3 March 2016 Accepted 9 March 2016
Objective: Backward locomotion in humans occurs during leisure, rehabilitation, and competitive sports. Little is known about its general biomechanical characteristics and how it affects lower extremity loading as well as muscle coordination. Thus, the purpose of this research was to analyze in-shoe plantar pressure patterns and lower extremity muscle activity patterns for backward compared to forward running. Methods: On a treadmill, nineteen runners performed forward running at their individually preferred speed, followed by backward running at 70% of their self-selected forward speed. In-shoe plantar pressures of nine foot regions and muscular activity of nine lower extremity muscles were recorded simultaneously over a one-minute interval. Backward and forward running variables were averaged over the accumulated steps and compared with Wilcoxon-signed rank tests (p < .05). Results: For backward compared to forward running, in-shoe plantar pressure distribution showed a load increase under metatarsal heads I and II, as well as under the medial midfoot. This was indicated by higher maximum forces and peak pressures, and by longer contact times. Muscle activity showed significantly higher mean amplitudes during backward running in the semitendinosus, rectus femoris, vastus lateralis, and gluteus medius during stance, and in the rectus femoris during swing phase, while significantly lower mean amplitudes were observed in the tibialis anterior during swing phase. Conclusion: Observations indicate plantar foot loading and muscle activity characteristics that are specific for the running direction. Thus, backward running may be used on purpose for certain rehabilitation tasks, aiming to strengthen respective lower extremity muscles. Furthermore, the findings are relevant for sport specific backward locomotion training. Finally, results provide an initial baseline for innovative athletic footwear development aiming to increase comfort and performance during backward running. ß 2016 Elsevier B.V. All rights reserved.
Keywords: Backward locomotion Plantar pressure patterns Gait line Pedobarography Electromyography
1. Introduction Forward locomotion of humans has evolved as the most effective strategy, due to the orientation of anatomical structures and vision. Initial biomechanical research revealed that backward running is characterized by smaller knee and hip joint ranges of motion, reflecting the restricted human anatomy regarding backward running [1]. These findings correspond to observations of shorter ground contact time and a shorter step length observed
* Corresponding author. Tel.: +49 251 835 2970; fax: +49 251 835 2993. E-mail address: [email protected] (D. Rosenbaum). http://dx.doi.org/10.1016/j.gaitpost.2016.03.009 0966-6362/ß 2016 Elsevier B.V. All rights reserved.
in backward running and sprinting [2,3]. Further biomechanical and neurophysiological differences were shown for backward compared to forward locomotion, among others addressing, the vertical ground reaction force component, lower extremity joint moments, barefoot plantar pressures, neuromuscular control strategies, and also trunk and head stabilization aspects [2,4–7]. Regarding biomechanical loading, a softer landing and harder takeoff was reported for backward running, combined with a lower efficiency of the stretch-shortening cycle of the involved muscle tendon units [8]. Due to the reversed nature of the touchdown and push-off characteristics compared to the normal heel–toe forward running style [9,10], the backward running impulse is directed more vertically during push-off than during the breaking phase of
136
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
ground contact [11]. Electromyography (EMG) measurements have illustrated differences in the lower extremity muscle activity between backward and forward running during stance and swing phase [12]. Backward locomotion is less cost-effective regarding human’s metabolic energy consumption than forward locomotion [13]. Various researchers examined preferred and energetically optimal transition speeds from backward walking to backward running [14,15], and compared transmission efficiency between backward and forward walking [16]. Regarding cardiovascular and metabolic aspects, backward walking and running require higher oxygen uptake and provoke higher heart rates compared to forward locomotion at the same absolute workload [13,17–19]. In general, backward running provokes an about 30% increased energy expenditure compared to forward running [17,19]. Backward locomotion is used in rehabilitation [2,20]. In clinical settings, backward oriented exercises are prescribed to improve coordination skills of patients, or when medical circumstances do not allow exposure to repetitive high loading of forward locomotion [20–22]. For instance, patellofemoral joint compression forces were reduced in backward compared to forward running [23]. While there is little need for backward oriented locomotion in daily life, it is more frequent in sports and leisure. In team sports like basketball, handball or soccer, backward oriented actions occur during defence movements to allow focussing on gamerelated circumstances. Racket sports like badminton or tennis require players to perform backward oriented steps and jumps. Thus, backward oriented acceleration, running and jumping mark important sport-specific requirements, and contribute to athletes’ performance [15]. Additionally, track-and-field backward sprinting and running are regarded as independent disciplines, with international competitions increasingly held and race distances ranging from 100 m to marathons. Backward locomotion appears very sensitive to training [3]. Already moderate training efforts resulted in improved acceleration, sprinting and running performance. Training effects were more pronounced with longer distances and predominantly achieved by an increased step length. It was further shown that locomotion speed is to a large extent specific for the individual, as faster forward runners were also faster backward runners. Despite the increasing relevance of backward oriented locomotion in leisure, rehabilitation and competitive sports, quantitative data on its specific biomechanical characteristics are scarce. For instance, general plantar pressure distribution and foot areas prone to high loading have not yet been described, even though foot loading analyses were shown to be of value for clinical [24,25], athletic [10], and footwear [26] aspects. The claim that backward running is effective for strengthening specific lower extremity muscles is only weakly supported. Therefore, the purpose of this research was to analyze in-shoe plantar pressures and muscle activity patterns of the lower extremity during backward compared to forward running. It was hypothesized that whole and regional plantar foot loading differ considerably in their magnitudes and roll-over characteristics between locomotion types. Further, it was hypothesized that backward running requires distinct EMG activity patterns of lower extremity muscles compared to forward running, reflecting its altered kinematic characteristics. 2. Methods 2.1. Participants Nineteen participants (eleven males, eight females; Mean (SD) age: 27.7 (7.5) years, height: 173.7 (11.6) cm, mass: 68.4 (14.2) kg,
BMI: 22.3 (2.1) kg/m2, shoe size: 41.6 3.0 EUR) were recruited for this laboratory study. They were volunteers from local running and triathlon clubs and provided written informed consent prior to participation in this research. Procedures of this research were approved by the local ethics committee prior to commencement. Participants were recreational or sub-elite heel–toe style runners when running forward, reflecting the predominant running style [9,10], and injury-free when tested. They did not have specific experience in backward running. 2.2. Instrumentation Backward and forward running was performed on a motordriven treadmill (Woodway GmbH, Weil am Rhein, Germany). Data collection comprised in-shoe plantar pressure distribution and muscle activity measurements of the lower extremity as characterized by EMG. In-shoe pressures of the left and right foot was measured by an in-sole system featuring 99 sensors per insole at a measurement frequency of 100 Hz (Pedar X, Novel, Munich, Germany). Recordings of in-shoe plantar pressures and EMG signals were synchronized to determine stance and swing phase intervals. EMG signals were collected by a surface electrode system at 2000 Hz (Noraxon Myosystem, Noraxon, Scottsdale, AZ, USA). Nine muscles of the right or left leg, randomized between participants, were measured: tibialis anterior, peroneus longus, soleus, lateral gastrocnemius, semitendinosus, biceps femoris, rectus femoris, vastus lateralis, gluteus medius. The SENIAM guidelines for bipolar surface EMG recordings were applied [27]. 2.3. Procedures Running was performed in participants’ own running shoes to avoid confounding factors due to shoe adaptation. Participants had sufficient time to familiarize with the forward and backward running tasks, which also served as warm-up. They performed forward running at individually preferred speed, and backward running at 70% of their self-selected forward speed, as backward running requires higher metabolic energy consumption than forward running [13]. Following the mounting of EMG instrumentation, participants’ individual referential amplitudes of maximum voluntary isometric contraction (MViC) of the selected muscles were taken [27,28]. Testing was performed by an experienced clinical technician for three to five seconds for each muscle. The treadmill was operated in the same direction during all testing, while runners switched their running orientation to perform the backward running task. All participants performed forward running first, followed by backward running, to avoid potential carry-over effects from unfamiliar backward running to familiar forward running. There was sufficient rest between the running tasks to avoid fatigue effects potentially caused by the forward running task. Duration of the forward and backward running task was 5 min each, with the fifth minute used for data collection. Hence, there was a 4 min adjustment period for participants to adopt a rhythmic running style. 2.4. Data processing and statistics The plantar aspect of the foot was divided into nine anatomical regions, using a modified version of the PRC mask [29]: medial rearfoot, lateral rearfoot, medial midfoot, lateral midfoot, metatarsal head I, metatarsal head II, metatarsal heads III–V, hallux, toes II–V. Contact area, contact time, maximum force, and peak pressures were analysed for each mask. Additionally, contact area, force-time integral, and pressure-time integral were analysed for the whole foot. For data processing of in-shoe pressures, the left or right foot of participants was randomly selected.
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
EMG signals were divided into stance and swing phase according to the synchronized plantar pressure information. Mean EMG amplitudes for stance and swing phase were determined for backward and forward running in relation to previously determined amplitudes of MViC. During data evaluation, in-shoe pressure data of ground contacts and mean EMG amplitudes of stance and swing phases of the 1 min data collection periods were first averaged for individual participants. Subsequently, means and standard deviations [Mean (SD)] were calculated across participants for all variables. In-shoe pressure and EMG variables were compared using nonparametric Wilcoxon-signed rank tests throughout, determining differences between locomotion types. Bonferroni correction was applied to pressure and EMG variables separately, resulting in adjusted a-levels of significance for pressure (.0013) and EMG (.0056) variables. Statistics were performed with Stat View 5.0, SAS Institute Inc., Cary, NC, USA.
3. Results Participants completed both running tasks without difficulties. Preferred forward running speed averaged to 2.64 (.43) m/s, and adjusted speed for backward running averaged to 1.85 (.29) m/s. Although speed was slower for backward running, participants used a higher number of steps during the 1 min measurement interval, resulting in 79 (23) steps for backward, and 72 (18) steps for forward running. The contact area of the whole foot during stance was reduced by 15% for backward [140.1 (25.5) cm2] compared to forward running [164.0 (23.3) cm2] (p = .0001). Maximum force of the whole foot did not differ between running directions (p = .3670), exhibiting similar forces for backward [1473 (353) N] and forward running [1495 (362) N]. The force-time integral of the whole foot differed between running directions, though not significantly (p = .0048), exhibiting a 5% reduction for backward [245.4 (66.6) N s] compared to forward running [258.5 (68.2) N s]. In contrast, the pressure-time integral of the whole foot exhibited a 20% increase (p = .0002) for backward [68.5 (13.0) kPa s] compared to forward running [57.2 (11.7) kPa s]. Plantar pressure parameters of the regional anatomical foot masks and their relative differences between backward and forward running are displayed in Table 1. During backward running, contact areas were significantly decreased in the rearfoot (medial: 43%, lateral: 49%), and midfoot regions (medial: 7%, lateral: 4%). Although not significantly different, contact areas tended to decrease also at the toe regions (hallux: 4%, toes II–V: 9%), whereas they were comparable in the three metatarsal head regions. Ground contact time during backward running was significantly decreased only for the lateral rearfoot region (30%). In the other foot masks ground contact times did not differ significantly. However, they tended to decrease in the medial rearfoot but tended to increase for midfoot, metatarsal head and toe regions during backward running. Maximum force during backward running was significantly increased under the metatarsal head I (30%), but significantly decreased in the rearfoot (medial: 83%, lateral: 85%), and toes 2–5 (34%). Plantar peak pressures during backward running showed a decrease for the medial (76%) and lateral (75%) rearfoot. In contrast, metatarsal head I and II, as well as the medial midfoot showed significantly increased plantar peak pressures (23–55%). Peak pressure differences of one participant illustrate the typical load shift from the rearfoot region towards the midfoot and metatarsal head regions observed during backward compared to forward running (Fig. 1). A sequential illustration of a selected participant’s pressure distribution over stance for backward compared to forward
137
Table 1 Plantar pressure variables and their relative differences (%) between backward and forward running, negative percentages indicate lower values for backward compared to forward running, positive percentages indicate respective higher values, significant p-values (after Bonferroni adjustment) marked in bold. Variable and foot region
Backward
Contact area [cm2] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
12.4 10.2 23.6 25.1 13.0 14.5 14.0 10.2 17.1
Contact time [ms] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
164.6 159.6 310.2 301.9 312.6 312.6 310.2 311.8 311.7
(93.9) (72.1) (70.1) (64.1) (70.1) (70.1) (70.0) (70.6) (71.1)
Maximum force [N] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
56.1 41.9 203.9 214.1 298.9 284.9 162.5 157.3 117.1
Peak pressure [kPa] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
52.8 50.8 242.7 164.8 366.1 308.7 211.9 299.8 178.3
(5.6) (5.7) (4.3) (4.2) (1.6) (1.8) (1.9) (1.5) (2.8)
Forward
21.8 19.7 25.3 26.2 12.9 14.5 14.0 10.6 18.7
p-Value
Difference [%]
(3.3) (3.6) (3.4) (3.6) (1.6) (1.8) (1.9) (1.7) (2.8)
.0002 .0002 .0004 .0002 n.a. n.a. n.a. .0251 .0015
43.0 48.5 7.0 4.3 .0 .0 .3 3.9 8.6
212.0 226.9 277.0 285.4 281.7 281.2 284.0 275.4 275.1
(66.4) (61.9) (39.2) (33.1) (35.4) (35.7) (34.3) (31.7) (33.9)
.0100 .0003 .0126 .1842 .0158 .0176 .0329 .0100 .0112
22.4 29.7 12.0 5.8 11.0 11.2 9.3 13.2 13.3
(48.7) (37.9) (91.6) (77.8) (66.8) (78.7) (55.3) (43.8) (30.1)
320.9 281.9 194.1 239.1 230.0 246.8 166.4 168.9 177.6
(107.0) (96.8) (73.4) (78.0) (36.9) (57.0) (55.7) (54.3) (78.3)
.0001 .0001 .2772 .0269 .0001 .0048 .3547 .1978 .0003
82.5 85.1 5.1 10.5 30.0 15.5 2.3 6.9 34.0
(22.0) (18.9) (107.2) (46.9) (100.0) (96.9) (57.0) (96.7) (46.4)
216.4 206.7 156.4 152.4 275.6 251.7 201.5 279.1 219.9
(64.0) (56.8) (37.5) (33.0) (57.9) (51.6) (48.3) (72.7) (57.0)
.0001 .0001 .0002 .0766 .0002 .0003 .2860 .1262 .0074
75.6 75.4 55.2 8.1 32.8 22.7 5.2 7.4 18.9
running indicates the different spatial and temporal peak pressure developments between locomotion types (Fig. 2). Relative muscle activity amplitudes during stance and swing phases are displayed in Table 2. These indicate significantly higher mean EMG amplitudes during backward compared to forward running in the semitendinosus, vastus lateralis, rectus femoris, and gluteus medius during stance (44–140%). Relative muscle activity amplitudes during swing phase indicate a significantly higher mean EMG amplitude during backward compared to forward running in the rectus femoris (50%), whereas mean EMG amplitude was reduced in the tibialis anterior (47%). An illustration of a selected participant’s averaged lower extremity EMG activity patterns for backward and forward running is displayed for the nine muscles over the gait cycle (Fig. 3). 4. Discussion This research compared plantar foot loading by in-shoe pressures and muscle activity patterns of the lower extremity by EMG during backward and forward running on a treadmill. The higher number of steps during backward running within consistent time intervals indicates a smaller step length during backward
138
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
Fig. 1. Left and right foot peak pressure differences [backward–forward (kPa)] of the maximum pressure pictures of one participant’s 1 min measurement cycle, indicating lower (blue) peak pressures at the rearfoot and higher (red) peak pressures at the forefoot during backward running. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 2. Peak pressure development (kPa) for one ground contact over stance of a participant’s left foot and centre of pressure paths (gait line) of the whole 1 min measurement cycles for backward (top) and forward (bottom) running. White dots indicate the centre of pressure for each loading phase. Gait lines of the repeated strides of the whole measurement appear more consistent in forward running and more variable in backward running.
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
139
Table 2 Mean EMG amplitudes [Mean (SD)] of lower extremity muscles during stance and swing phase as percentage (%) of maximum voluntary isometric contraction and relative differences between backward and forward mean EMG amplitudes; negative difference percentages indicate lower EMG values for backward compared to forward running, positive percentages indicate respectively higher values, significant p-values (after Bonferroni adjustment) marked in bold. Muscle
Tibialis anterior Peroneus longus Soleus Gastrocnemius lateralis Semitendinosus Biceps Femoris Rectus femoris Vastus lateralis Gluteus medius
Stance phase
Swing phase
Backward
Forward
p-Value
Difference [%]
Backward
Forward
p-Value
37 106 101 115 54 37 115 193 135
31 113 108 122 35 48 48 130 94
.0840 .3560 .2461 .6874 .0036 .1359 .0004 .0004 .0026
19.4 6.2 6.5 5.7 54.3 22.9 139.6 48.5 43.6
56 41 31 27 54 35 39 44 52
105 32 20 19 56 38 26 42 46
.0003 .0759 .0148 .3088 .9058 .4925 .0042 .7946 .4348
(13) (55) (25) (37) (37) (17) (46) (69) (78)
(10) (55) (27) (52) (28) (31) (16) (43) (52)
(21) (23) (14) (17) (35) (14) (20) (20) (31)
(46) (10) (6) (8) (33) (17) (13) (17) (20)
Difference [%] 46.7 28.1 55 42.1 3.6 7.9 50 4.8 13
Fig. 3. Direct comparison of the averaged EMG activity (%) patterns (normalized to 100% gait cycle; 0/100% = initial contact) of all steps of the whole 1 min measurement cycle for one participant during backward running (‘bw’ in grey) and forward running (‘fw’ in black); TA = tibialis anterior, PL = peroneus longus, SO = soleus, LG = lateral gastrocnemius, ST = semitendinosus, BF = biceps femoris, RF = rectus femoris, VL = vastus lateralis, GM = gluteus medius.
running and suggests shorter ground contact times, corresponding to earlier findings [1–3]. This holds especially true as backward running speed in this research was reduced to 70% of forward running speed. In-shoe plantar loading during backward running showed different loading characteristics of the whole and regional foot areas compared to forward running, confirming our hypothesis. Maximum forces under the whole foot tended to be moderately reduced during backward running, which is attributed to the lower speed set, and the higher step count observed. In contrast, pressure loading under the whole foot was considerably increased during backward running, a mechanism triggered by a considerably reduced contact area of the whole foot exhibited by our runners.
Such increased pressure loading during backward running marks a factor to consider regarding running comfort and also regarding the potential development of overuse injuries triggered during extensive backward running training regimens. The reduced plantar contact area during backward running is predominantly induced by lesser involvement of the rearfoot and midfoot. In contrast, the metatarsal head regions as the main loading zones are used to a similar extent in both running directions. This loading strategy is reflected by regional peak pressures, which are markedly reduced at the rearfoot. Higher peak pressures during backward running are predominantly seen under the first and second metatarsal head, as well as under the medial midfoot region. Thus, backward running predominantly causes a
140
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
load increase at the medial midfoot to forefoot, shown by peak pressures and additionally illustrated by maximum forces exhibited at these regions. These findings appear useful regarding considerations towards specific footwear for backward running, aiming to increase running comfort. Pressure patterns during backward running reflect the initial forefoot contact in comparison to the initial heel contact commonly seen in forward running. Thus, backward running, compared to heel–toe forward running, provokes a forefootdominated stance pattern. This is additionally illustrated by a tendency of increased loading times in the midfoot, forefoot and toe regions, and reduced loading times in the rearfoot regions, especially regarding their lateral aspect. In contrast to the typical heel–toe forward running style, backward running is characterized by a forefoot-midfoot-forefoot pattern, providing a more detailed illustration of foot-ground interaction than the previously referred to toe–heel pattern during backward running [1,8] (Fig. 3). These different roll-over characteristics found during stance correspond to the reported softer touch down and harder take-off during backward running [8]. Thereby, the softer touchdown during backward running is mediated by more flexible forefoot structures compared to the rather rigid heel structure, mediating initial contact during forward running. Muscle activity patterns differed between backward and forward running for some lower extremity muscles, confirming our hypothesis. Higher muscle activity was observed during backward running despite backward running speed was reduced in this research. The reversal of the movement direction affected mean EMG amplitudes during stance, especially in those muscles responsible for knee (rectus femoris, vastus lateralis), and hip extension (semitendinosus, gluteus medius), supporting the earlier notion that backward running training can increase knee extensor strength [2]. These higher muscular efforts contributing to knee and hip extension correspond to the more vertical impulse found in backward running [11]. In contrast, shank muscles did not show differences in mean muscle activity amplitudes during stance. Thus, the increase of muscle activity during stance predominantly observed in the strong thigh and hip muscles seems responsible for the higher metabolic costs of backward running [13,17–19]. During swing phase some shank muscles revealed different activation patterns during backward compared to forward running, confirming our hypothesis. Whereas the tibialis anterior, as the main dorsiflexor muscle, showed decreased mean EMG amplitude, the soleus and gastrocnemius lateralis, as plantarflexor muscles, indicate increased mean EMG amplitudes during backward running, though for the latter muscles differences did not reach statistical significance despite rather high mean value differences. This pattern reflects the different preparation of touchdown between locomotion types. During heel–toe forward running the tibialis anterior is responsible for lifting the forefoot to accommodate the heel touchdown, while the soleus as plantar flexor muscle is relatively relaxed. During backward running foremost the soleus and to a lesser degree also the gastrocnemius lateralis contribute to the voluntary plantar flexion of the foot accommodating the forefoot touchdown, while the tibialis anterior remains relatively relaxed. The increased swing phase EMG amplitude of the rectus femoris during backward running may indicate hip flexion before touchdown, contributing to the softer touchdown reported [8]. Our findings of distinctly different muscle activation patterns of selected muscles support the notion that backward running has the potential to train selective muscles in different ways forward running does. Limitations of this research are acknowledged. Although backward running speed was reduced compared to forward running speed similar metabolic costs could not be guaranteed
for each participant during the running tasks as individual backward running coordination levels of the novice backward runners may have differed. As our runners and triathletes did not have specific experience in backward running, results cannot be generalized to team sport athletes, who are exposed to backward running due to the multi-directional nature of their sports, or to athletes with solid backward running practice [3]. It remains to be seen whether backward running exercise results in more efficient muscle activity patterns and lesser increase in regional plantar loading due to altered coordination skills of athletes. Future research should investigate whether athletes having acquired automated backward running styles evoke modified plantar pressure patterns due to more economical coordination [3,30]. 5. Conclusion Backward running exhibits specific foot roll-over characteristics initiating at the forefoot during touchdown, progressing to the midfoot/rearfoot during midstance, and then progressing back to the forefoot to generate the push-off. This forefoot dominant locomotion style is characterized by higher loading of the forefoot compared to the normal heel–toe forward running style. Muscle activity was increased for backward compared to forward running, predominantly during stance and foremost at the thigh and hip muscles, although running speed was adjusted to account for higher metabolic costs during backward running. Future research should investigate whether training programmes for backward running decrease required muscle activity and reduce the high forefoot loading. Acknowledgements The authors thank Helga Raape for assistance during EMG measurements of MViC and Dieter Klein for technical support of this research. Conflict of interest statement The authors have no financial and personal relationships with other people or organizations that could inappropriately influence their work. References [1] Bates BT, Morrison E, Hamill J. Differences between forward and backward running. In: Adrian M, Deutsch H, editors. Proceedings 1984 Olympic Scientific Congress: Biomechanics, July 19–26. Eugene, OR, USA: University of Oregon Microform Publications; 1984. p. 127–35. [2] Threlkeld AJ, Horn TS, Wojtowicz GM, Rooney JG, Shapiro R. Kinematics, ground reaction force, and muscle balance produced by backward running. J. Orthop. Sports Phys. Ther. 1989;11:56–63. [3] Schubert C, Brauner T, Eckardt D, Sterzing T. Trainierbarkeit des Ru¨ckwa¨rtslaufens. 8. Gemeinsames Symposium der dvs-Sektionen Biomechanik, Sportmotorik und Trainingswissenschaft, Hamburg, Deutschland. In: Mattes, Wollesen, editors. dvs Band 204: Bewegung und Leistung—Sport, Gesundheit & Alter. Hamburg, Deutschland: Feldhaus Verlag Edition Czwalina; 2010. p. 44. [4] Nadeau S, Amblard B, Mesure S, Bourbonnais D. Head and trunk stabilization strategies during forward and backward walking in healthy adults. Gait Posture 2003;181:134–42. [5] Rosenbaum D, Schubert C, Sterzing T. Gait speed and plantar pressure pattern differences in forward and backward walking with eyes open and closed. In: Van Sint JS, et al., editors. Proceedings of the 23. Congress of the International Society of Biomechanics. Belgium: Vrije Universiteit Brussel; 2011. p. 190–1. [6] Jansen K, De Groote F, Massaad F, Meyns P, Duysens J, Jonkers I. Similar muscles contribute to horizontal and vertical acceleration of center of mass in forward and backward walking: Implications for neural control. J. Neurophysiol. 2012;107:3385–96. [7] DeVita P, Stribling J. Lower extremity joint kinetics and energetics during backward running. Med. Sci. Sports Exerc. 1991;23:602–10. [8] Cavagna GA, Legramandi MA, La Torre A. Running backwards: soft landing— hard takeoff, a less efficient rebound. Proc. Biol. Sci. 2011;278:339–46.
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141 [9] Larson P, Higgins E, Kaminski J, Decker T, Preble J, Lyons D, McIntyre K, Normile A. Foot strike patterns of recreational and sub-elite runners in a long-distance road race. J. Sports Sci. 2011;29:1665–73. [10] Breine B, Malcolm P, Frederick EC, de Clercq D. Relationship between running speed and initial foot contact patterns. Med. Sci. Sports Exerc. 2014;46: 1595–603. [11] Cavagna GA, Legramandi MA, La Torre A. An analysis of the rebound of the body in backward human running. J. Exp. Biol. 2012;215:75–84. [12] Flynn TW, Soutas-Little RW. Mechanical power and muscle action during forward and backward running. J. Orthop. Sports Phys. Ther. 1993;17:108–12. [13] Chaloupka EC, Kang J, Mastrangelo MA, Donnelly MC. Cardiorespiratory and metabolic responses during forward and backward walking. J. Orthop. Sports Phys. Ther. 1997;25:302–6. [14] Terblanche E, Cloete WA, du Plessis PAL, Sadie JN, Strauss A, Unger M. The metabolic transition speed between backward walking and running. Eur. J. Appl. Physiol. 2003;90:520–5. [15] Hreljac A, Imamura R, Escamilla RF, Casebolt J, Sision M. Preferred and energetically optimal transition speeds during backward human locomotion. J. Sports Sci. Med. 2005;4:446–54. [16] Minetti AE, Ardigo` LP. The transmission efficiency of backward walking at different gradients. Pflugers Arch. 2001;442:542–6. [17] Flynn TW, Connery SM, Smutok MA, Zeballos RJ, Weisman IM. Comparison of cardiopulmonary responses to forward and backward walking and running. Med. Sci. Sports Exerc. 1994;26:89–94. [18] Williford HN, Olson MS, Gauger S, Duey WJ, Blessing DL. Cardiovascular and metabolic costs of forward, backward, and lateral motion. Med. Sci. Sports Exerc. 1998;30:1419–23. [19] Wright S, Weyand PG. The application of ground forces explains the energetic cost of running backward and forward. J. Exp. Biol. 2001;204:1805–15.
141
[20] Zampeli F, Moraiti CO, Xergia S, Tsiaras VA, Stergiou N, Georgoulis AD. Strideto-stride variability is altered during backward walking in anterior cruciate ligament deficient patients. Clin. Biomech. 2010;25:1037–41. [21] Heath EM, Blackwell JR, Baker UC, Smith DR, Kornatz KW. Backward walking practice decreases oxygen uptake, heart rate and ratings of perceived exertion. Phys. Ther. Sport 2001;2:171–7. [22] Hooper TL, Dunn DM, Erick Props J, Bruce BA, Sawyer SF, Daniel JA. The effects of graded forward and backward walking on heart rate and oxygen consumption. J. Orthop. Sports Phys. Ther. 2004;34:65–71. [23] Roos PE, Barton N, van Deursen RWM. Patellofemoral joint compression forces in backward and forward running. J. Biomech. 2012;45:1656–60. [24] Hughes J. The clinical use of pedobarography. Acta Orthop. Belg. 1993;59:10–6. [25] Rosenbaum D, Becker HP. Plantar pressure distribution measurements. Technical background and clinical evaluation. Foot Ankle Surg. 1997;3:1–14. [26] Sterzing T, Custoza G, Ding R, Cheung JT-M. Segmented midsole hardness in the midfoot to forefoot region of running shoes alters subjective perception and biomechanics during heel–toe running revealing potential to enhance footwear. Footwear Sci. 2015;7:63–79. [27] Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J. Electromyogr. Kinesiol. 2000;10:361–74. [28] Weist R, Eils E, Rosenbaum D. The influence of muscle fatigue on electromyogram and plantar pressure patterns as an explanation for the incidence of metatarsal stress fractures. Am. J. Sports Med. 2004;32:1893–8. [29] Cavanagh PR, Ulbrecht JS. Clinical plantar pressure measurement in diabetes: rationale and methodology. Foot 1994;4:123–35. [30] Mehdizadeh S, Arshi AR, Davids K. Quantifying coordination and coordination variability in backward versus forward running: implications for control of motion. Gait Posture 2015;42:172–7.
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Full length article
In-shoe plantar pressure distribution and lower extremity muscle activity patterns of backward compared to forward running on a treadmill Thorsten Sterzing a,c, Clivia Frommhold b,c, Dieter Rosenbaum d,* a
Li Ning Sports Science Research Centre, Beijing, China Department Medical Affairs, Bauerfeind AG, Zeulenroda, Germany Department of Human Locomotion, Chemnitz University of Technology, Chemnitz, Germany d Movement Analysis Lab, Institute for Experimental Musculoskeletal Medicine, University Hospital Mu¨nster, Mu¨nster, Germany b c
A R T I C L E I N F O
A B S T R A C T
Article history: Received 23 September 2015 Received in revised form 3 March 2016 Accepted 9 March 2016
Objective: Backward locomotion in humans occurs during leisure, rehabilitation, and competitive sports. Little is known about its general biomechanical characteristics and how it affects lower extremity loading as well as muscle coordination. Thus, the purpose of this research was to analyze in-shoe plantar pressure patterns and lower extremity muscle activity patterns for backward compared to forward running. Methods: On a treadmill, nineteen runners performed forward running at their individually preferred speed, followed by backward running at 70% of their self-selected forward speed. In-shoe plantar pressures of nine foot regions and muscular activity of nine lower extremity muscles were recorded simultaneously over a one-minute interval. Backward and forward running variables were averaged over the accumulated steps and compared with Wilcoxon-signed rank tests (p < .05). Results: For backward compared to forward running, in-shoe plantar pressure distribution showed a load increase under metatarsal heads I and II, as well as under the medial midfoot. This was indicated by higher maximum forces and peak pressures, and by longer contact times. Muscle activity showed significantly higher mean amplitudes during backward running in the semitendinosus, rectus femoris, vastus lateralis, and gluteus medius during stance, and in the rectus femoris during swing phase, while significantly lower mean amplitudes were observed in the tibialis anterior during swing phase. Conclusion: Observations indicate plantar foot loading and muscle activity characteristics that are specific for the running direction. Thus, backward running may be used on purpose for certain rehabilitation tasks, aiming to strengthen respective lower extremity muscles. Furthermore, the findings are relevant for sport specific backward locomotion training. Finally, results provide an initial baseline for innovative athletic footwear development aiming to increase comfort and performance during backward running. ß 2016 Elsevier B.V. All rights reserved.
Keywords: Backward locomotion Plantar pressure patterns Gait line Pedobarography Electromyography
1. Introduction Forward locomotion of humans has evolved as the most effective strategy, due to the orientation of anatomical structures and vision. Initial biomechanical research revealed that backward running is characterized by smaller knee and hip joint ranges of motion, reflecting the restricted human anatomy regarding backward running [1]. These findings correspond to observations of shorter ground contact time and a shorter step length observed
* Corresponding author. Tel.: +49 251 835 2970; fax: +49 251 835 2993. E-mail address: [email protected] (D. Rosenbaum). http://dx.doi.org/10.1016/j.gaitpost.2016.03.009 0966-6362/ß 2016 Elsevier B.V. All rights reserved.
in backward running and sprinting [2,3]. Further biomechanical and neurophysiological differences were shown for backward compared to forward locomotion, among others addressing, the vertical ground reaction force component, lower extremity joint moments, barefoot plantar pressures, neuromuscular control strategies, and also trunk and head stabilization aspects [2,4–7]. Regarding biomechanical loading, a softer landing and harder takeoff was reported for backward running, combined with a lower efficiency of the stretch-shortening cycle of the involved muscle tendon units [8]. Due to the reversed nature of the touchdown and push-off characteristics compared to the normal heel–toe forward running style [9,10], the backward running impulse is directed more vertically during push-off than during the breaking phase of
136
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
ground contact [11]. Electromyography (EMG) measurements have illustrated differences in the lower extremity muscle activity between backward and forward running during stance and swing phase [12]. Backward locomotion is less cost-effective regarding human’s metabolic energy consumption than forward locomotion [13]. Various researchers examined preferred and energetically optimal transition speeds from backward walking to backward running [14,15], and compared transmission efficiency between backward and forward walking [16]. Regarding cardiovascular and metabolic aspects, backward walking and running require higher oxygen uptake and provoke higher heart rates compared to forward locomotion at the same absolute workload [13,17–19]. In general, backward running provokes an about 30% increased energy expenditure compared to forward running [17,19]. Backward locomotion is used in rehabilitation [2,20]. In clinical settings, backward oriented exercises are prescribed to improve coordination skills of patients, or when medical circumstances do not allow exposure to repetitive high loading of forward locomotion [20–22]. For instance, patellofemoral joint compression forces were reduced in backward compared to forward running [23]. While there is little need for backward oriented locomotion in daily life, it is more frequent in sports and leisure. In team sports like basketball, handball or soccer, backward oriented actions occur during defence movements to allow focussing on gamerelated circumstances. Racket sports like badminton or tennis require players to perform backward oriented steps and jumps. Thus, backward oriented acceleration, running and jumping mark important sport-specific requirements, and contribute to athletes’ performance [15]. Additionally, track-and-field backward sprinting and running are regarded as independent disciplines, with international competitions increasingly held and race distances ranging from 100 m to marathons. Backward locomotion appears very sensitive to training [3]. Already moderate training efforts resulted in improved acceleration, sprinting and running performance. Training effects were more pronounced with longer distances and predominantly achieved by an increased step length. It was further shown that locomotion speed is to a large extent specific for the individual, as faster forward runners were also faster backward runners. Despite the increasing relevance of backward oriented locomotion in leisure, rehabilitation and competitive sports, quantitative data on its specific biomechanical characteristics are scarce. For instance, general plantar pressure distribution and foot areas prone to high loading have not yet been described, even though foot loading analyses were shown to be of value for clinical [24,25], athletic [10], and footwear [26] aspects. The claim that backward running is effective for strengthening specific lower extremity muscles is only weakly supported. Therefore, the purpose of this research was to analyze in-shoe plantar pressures and muscle activity patterns of the lower extremity during backward compared to forward running. It was hypothesized that whole and regional plantar foot loading differ considerably in their magnitudes and roll-over characteristics between locomotion types. Further, it was hypothesized that backward running requires distinct EMG activity patterns of lower extremity muscles compared to forward running, reflecting its altered kinematic characteristics. 2. Methods 2.1. Participants Nineteen participants (eleven males, eight females; Mean (SD) age: 27.7 (7.5) years, height: 173.7 (11.6) cm, mass: 68.4 (14.2) kg,
BMI: 22.3 (2.1) kg/m2, shoe size: 41.6 3.0 EUR) were recruited for this laboratory study. They were volunteers from local running and triathlon clubs and provided written informed consent prior to participation in this research. Procedures of this research were approved by the local ethics committee prior to commencement. Participants were recreational or sub-elite heel–toe style runners when running forward, reflecting the predominant running style [9,10], and injury-free when tested. They did not have specific experience in backward running. 2.2. Instrumentation Backward and forward running was performed on a motordriven treadmill (Woodway GmbH, Weil am Rhein, Germany). Data collection comprised in-shoe plantar pressure distribution and muscle activity measurements of the lower extremity as characterized by EMG. In-shoe pressures of the left and right foot was measured by an in-sole system featuring 99 sensors per insole at a measurement frequency of 100 Hz (Pedar X, Novel, Munich, Germany). Recordings of in-shoe plantar pressures and EMG signals were synchronized to determine stance and swing phase intervals. EMG signals were collected by a surface electrode system at 2000 Hz (Noraxon Myosystem, Noraxon, Scottsdale, AZ, USA). Nine muscles of the right or left leg, randomized between participants, were measured: tibialis anterior, peroneus longus, soleus, lateral gastrocnemius, semitendinosus, biceps femoris, rectus femoris, vastus lateralis, gluteus medius. The SENIAM guidelines for bipolar surface EMG recordings were applied [27]. 2.3. Procedures Running was performed in participants’ own running shoes to avoid confounding factors due to shoe adaptation. Participants had sufficient time to familiarize with the forward and backward running tasks, which also served as warm-up. They performed forward running at individually preferred speed, and backward running at 70% of their self-selected forward speed, as backward running requires higher metabolic energy consumption than forward running [13]. Following the mounting of EMG instrumentation, participants’ individual referential amplitudes of maximum voluntary isometric contraction (MViC) of the selected muscles were taken [27,28]. Testing was performed by an experienced clinical technician for three to five seconds for each muscle. The treadmill was operated in the same direction during all testing, while runners switched their running orientation to perform the backward running task. All participants performed forward running first, followed by backward running, to avoid potential carry-over effects from unfamiliar backward running to familiar forward running. There was sufficient rest between the running tasks to avoid fatigue effects potentially caused by the forward running task. Duration of the forward and backward running task was 5 min each, with the fifth minute used for data collection. Hence, there was a 4 min adjustment period for participants to adopt a rhythmic running style. 2.4. Data processing and statistics The plantar aspect of the foot was divided into nine anatomical regions, using a modified version of the PRC mask [29]: medial rearfoot, lateral rearfoot, medial midfoot, lateral midfoot, metatarsal head I, metatarsal head II, metatarsal heads III–V, hallux, toes II–V. Contact area, contact time, maximum force, and peak pressures were analysed for each mask. Additionally, contact area, force-time integral, and pressure-time integral were analysed for the whole foot. For data processing of in-shoe pressures, the left or right foot of participants was randomly selected.
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
EMG signals were divided into stance and swing phase according to the synchronized plantar pressure information. Mean EMG amplitudes for stance and swing phase were determined for backward and forward running in relation to previously determined amplitudes of MViC. During data evaluation, in-shoe pressure data of ground contacts and mean EMG amplitudes of stance and swing phases of the 1 min data collection periods were first averaged for individual participants. Subsequently, means and standard deviations [Mean (SD)] were calculated across participants for all variables. In-shoe pressure and EMG variables were compared using nonparametric Wilcoxon-signed rank tests throughout, determining differences between locomotion types. Bonferroni correction was applied to pressure and EMG variables separately, resulting in adjusted a-levels of significance for pressure (.0013) and EMG (.0056) variables. Statistics were performed with Stat View 5.0, SAS Institute Inc., Cary, NC, USA.
3. Results Participants completed both running tasks without difficulties. Preferred forward running speed averaged to 2.64 (.43) m/s, and adjusted speed for backward running averaged to 1.85 (.29) m/s. Although speed was slower for backward running, participants used a higher number of steps during the 1 min measurement interval, resulting in 79 (23) steps for backward, and 72 (18) steps for forward running. The contact area of the whole foot during stance was reduced by 15% for backward [140.1 (25.5) cm2] compared to forward running [164.0 (23.3) cm2] (p = .0001). Maximum force of the whole foot did not differ between running directions (p = .3670), exhibiting similar forces for backward [1473 (353) N] and forward running [1495 (362) N]. The force-time integral of the whole foot differed between running directions, though not significantly (p = .0048), exhibiting a 5% reduction for backward [245.4 (66.6) N s] compared to forward running [258.5 (68.2) N s]. In contrast, the pressure-time integral of the whole foot exhibited a 20% increase (p = .0002) for backward [68.5 (13.0) kPa s] compared to forward running [57.2 (11.7) kPa s]. Plantar pressure parameters of the regional anatomical foot masks and their relative differences between backward and forward running are displayed in Table 1. During backward running, contact areas were significantly decreased in the rearfoot (medial: 43%, lateral: 49%), and midfoot regions (medial: 7%, lateral: 4%). Although not significantly different, contact areas tended to decrease also at the toe regions (hallux: 4%, toes II–V: 9%), whereas they were comparable in the three metatarsal head regions. Ground contact time during backward running was significantly decreased only for the lateral rearfoot region (30%). In the other foot masks ground contact times did not differ significantly. However, they tended to decrease in the medial rearfoot but tended to increase for midfoot, metatarsal head and toe regions during backward running. Maximum force during backward running was significantly increased under the metatarsal head I (30%), but significantly decreased in the rearfoot (medial: 83%, lateral: 85%), and toes 2–5 (34%). Plantar peak pressures during backward running showed a decrease for the medial (76%) and lateral (75%) rearfoot. In contrast, metatarsal head I and II, as well as the medial midfoot showed significantly increased plantar peak pressures (23–55%). Peak pressure differences of one participant illustrate the typical load shift from the rearfoot region towards the midfoot and metatarsal head regions observed during backward compared to forward running (Fig. 1). A sequential illustration of a selected participant’s pressure distribution over stance for backward compared to forward
137
Table 1 Plantar pressure variables and their relative differences (%) between backward and forward running, negative percentages indicate lower values for backward compared to forward running, positive percentages indicate respective higher values, significant p-values (after Bonferroni adjustment) marked in bold. Variable and foot region
Backward
Contact area [cm2] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
12.4 10.2 23.6 25.1 13.0 14.5 14.0 10.2 17.1
Contact time [ms] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
164.6 159.6 310.2 301.9 312.6 312.6 310.2 311.8 311.7
(93.9) (72.1) (70.1) (64.1) (70.1) (70.1) (70.0) (70.6) (71.1)
Maximum force [N] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
56.1 41.9 203.9 214.1 298.9 284.9 162.5 157.3 117.1
Peak pressure [kPa] Medial rearfoot Lateral rearfoot Medial midfoot Lateral midfoot Metatarsal head I Metatarsal head II Metatarsal head III–V Hallux Toes II–V
52.8 50.8 242.7 164.8 366.1 308.7 211.9 299.8 178.3
(5.6) (5.7) (4.3) (4.2) (1.6) (1.8) (1.9) (1.5) (2.8)
Forward
21.8 19.7 25.3 26.2 12.9 14.5 14.0 10.6 18.7
p-Value
Difference [%]
(3.3) (3.6) (3.4) (3.6) (1.6) (1.8) (1.9) (1.7) (2.8)
.0002 .0002 .0004 .0002 n.a. n.a. n.a. .0251 .0015
43.0 48.5 7.0 4.3 .0 .0 .3 3.9 8.6
212.0 226.9 277.0 285.4 281.7 281.2 284.0 275.4 275.1
(66.4) (61.9) (39.2) (33.1) (35.4) (35.7) (34.3) (31.7) (33.9)
.0100 .0003 .0126 .1842 .0158 .0176 .0329 .0100 .0112
22.4 29.7 12.0 5.8 11.0 11.2 9.3 13.2 13.3
(48.7) (37.9) (91.6) (77.8) (66.8) (78.7) (55.3) (43.8) (30.1)
320.9 281.9 194.1 239.1 230.0 246.8 166.4 168.9 177.6
(107.0) (96.8) (73.4) (78.0) (36.9) (57.0) (55.7) (54.3) (78.3)
.0001 .0001 .2772 .0269 .0001 .0048 .3547 .1978 .0003
82.5 85.1 5.1 10.5 30.0 15.5 2.3 6.9 34.0
(22.0) (18.9) (107.2) (46.9) (100.0) (96.9) (57.0) (96.7) (46.4)
216.4 206.7 156.4 152.4 275.6 251.7 201.5 279.1 219.9
(64.0) (56.8) (37.5) (33.0) (57.9) (51.6) (48.3) (72.7) (57.0)
.0001 .0001 .0002 .0766 .0002 .0003 .2860 .1262 .0074
75.6 75.4 55.2 8.1 32.8 22.7 5.2 7.4 18.9
running indicates the different spatial and temporal peak pressure developments between locomotion types (Fig. 2). Relative muscle activity amplitudes during stance and swing phases are displayed in Table 2. These indicate significantly higher mean EMG amplitudes during backward compared to forward running in the semitendinosus, vastus lateralis, rectus femoris, and gluteus medius during stance (44–140%). Relative muscle activity amplitudes during swing phase indicate a significantly higher mean EMG amplitude during backward compared to forward running in the rectus femoris (50%), whereas mean EMG amplitude was reduced in the tibialis anterior (47%). An illustration of a selected participant’s averaged lower extremity EMG activity patterns for backward and forward running is displayed for the nine muscles over the gait cycle (Fig. 3). 4. Discussion This research compared plantar foot loading by in-shoe pressures and muscle activity patterns of the lower extremity by EMG during backward and forward running on a treadmill. The higher number of steps during backward running within consistent time intervals indicates a smaller step length during backward
138
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
Fig. 1. Left and right foot peak pressure differences [backward–forward (kPa)] of the maximum pressure pictures of one participant’s 1 min measurement cycle, indicating lower (blue) peak pressures at the rearfoot and higher (red) peak pressures at the forefoot during backward running. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 2. Peak pressure development (kPa) for one ground contact over stance of a participant’s left foot and centre of pressure paths (gait line) of the whole 1 min measurement cycles for backward (top) and forward (bottom) running. White dots indicate the centre of pressure for each loading phase. Gait lines of the repeated strides of the whole measurement appear more consistent in forward running and more variable in backward running.
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
139
Table 2 Mean EMG amplitudes [Mean (SD)] of lower extremity muscles during stance and swing phase as percentage (%) of maximum voluntary isometric contraction and relative differences between backward and forward mean EMG amplitudes; negative difference percentages indicate lower EMG values for backward compared to forward running, positive percentages indicate respectively higher values, significant p-values (after Bonferroni adjustment) marked in bold. Muscle
Tibialis anterior Peroneus longus Soleus Gastrocnemius lateralis Semitendinosus Biceps Femoris Rectus femoris Vastus lateralis Gluteus medius
Stance phase
Swing phase
Backward
Forward
p-Value
Difference [%]
Backward
Forward
p-Value
37 106 101 115 54 37 115 193 135
31 113 108 122 35 48 48 130 94
.0840 .3560 .2461 .6874 .0036 .1359 .0004 .0004 .0026
19.4 6.2 6.5 5.7 54.3 22.9 139.6 48.5 43.6
56 41 31 27 54 35 39 44 52
105 32 20 19 56 38 26 42 46
.0003 .0759 .0148 .3088 .9058 .4925 .0042 .7946 .4348
(13) (55) (25) (37) (37) (17) (46) (69) (78)
(10) (55) (27) (52) (28) (31) (16) (43) (52)
(21) (23) (14) (17) (35) (14) (20) (20) (31)
(46) (10) (6) (8) (33) (17) (13) (17) (20)
Difference [%] 46.7 28.1 55 42.1 3.6 7.9 50 4.8 13
Fig. 3. Direct comparison of the averaged EMG activity (%) patterns (normalized to 100% gait cycle; 0/100% = initial contact) of all steps of the whole 1 min measurement cycle for one participant during backward running (‘bw’ in grey) and forward running (‘fw’ in black); TA = tibialis anterior, PL = peroneus longus, SO = soleus, LG = lateral gastrocnemius, ST = semitendinosus, BF = biceps femoris, RF = rectus femoris, VL = vastus lateralis, GM = gluteus medius.
running and suggests shorter ground contact times, corresponding to earlier findings [1–3]. This holds especially true as backward running speed in this research was reduced to 70% of forward running speed. In-shoe plantar loading during backward running showed different loading characteristics of the whole and regional foot areas compared to forward running, confirming our hypothesis. Maximum forces under the whole foot tended to be moderately reduced during backward running, which is attributed to the lower speed set, and the higher step count observed. In contrast, pressure loading under the whole foot was considerably increased during backward running, a mechanism triggered by a considerably reduced contact area of the whole foot exhibited by our runners.
Such increased pressure loading during backward running marks a factor to consider regarding running comfort and also regarding the potential development of overuse injuries triggered during extensive backward running training regimens. The reduced plantar contact area during backward running is predominantly induced by lesser involvement of the rearfoot and midfoot. In contrast, the metatarsal head regions as the main loading zones are used to a similar extent in both running directions. This loading strategy is reflected by regional peak pressures, which are markedly reduced at the rearfoot. Higher peak pressures during backward running are predominantly seen under the first and second metatarsal head, as well as under the medial midfoot region. Thus, backward running predominantly causes a
140
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141
load increase at the medial midfoot to forefoot, shown by peak pressures and additionally illustrated by maximum forces exhibited at these regions. These findings appear useful regarding considerations towards specific footwear for backward running, aiming to increase running comfort. Pressure patterns during backward running reflect the initial forefoot contact in comparison to the initial heel contact commonly seen in forward running. Thus, backward running, compared to heel–toe forward running, provokes a forefootdominated stance pattern. This is additionally illustrated by a tendency of increased loading times in the midfoot, forefoot and toe regions, and reduced loading times in the rearfoot regions, especially regarding their lateral aspect. In contrast to the typical heel–toe forward running style, backward running is characterized by a forefoot-midfoot-forefoot pattern, providing a more detailed illustration of foot-ground interaction than the previously referred to toe–heel pattern during backward running [1,8] (Fig. 3). These different roll-over characteristics found during stance correspond to the reported softer touch down and harder take-off during backward running [8]. Thereby, the softer touchdown during backward running is mediated by more flexible forefoot structures compared to the rather rigid heel structure, mediating initial contact during forward running. Muscle activity patterns differed between backward and forward running for some lower extremity muscles, confirming our hypothesis. Higher muscle activity was observed during backward running despite backward running speed was reduced in this research. The reversal of the movement direction affected mean EMG amplitudes during stance, especially in those muscles responsible for knee (rectus femoris, vastus lateralis), and hip extension (semitendinosus, gluteus medius), supporting the earlier notion that backward running training can increase knee extensor strength [2]. These higher muscular efforts contributing to knee and hip extension correspond to the more vertical impulse found in backward running [11]. In contrast, shank muscles did not show differences in mean muscle activity amplitudes during stance. Thus, the increase of muscle activity during stance predominantly observed in the strong thigh and hip muscles seems responsible for the higher metabolic costs of backward running [13,17–19]. During swing phase some shank muscles revealed different activation patterns during backward compared to forward running, confirming our hypothesis. Whereas the tibialis anterior, as the main dorsiflexor muscle, showed decreased mean EMG amplitude, the soleus and gastrocnemius lateralis, as plantarflexor muscles, indicate increased mean EMG amplitudes during backward running, though for the latter muscles differences did not reach statistical significance despite rather high mean value differences. This pattern reflects the different preparation of touchdown between locomotion types. During heel–toe forward running the tibialis anterior is responsible for lifting the forefoot to accommodate the heel touchdown, while the soleus as plantar flexor muscle is relatively relaxed. During backward running foremost the soleus and to a lesser degree also the gastrocnemius lateralis contribute to the voluntary plantar flexion of the foot accommodating the forefoot touchdown, while the tibialis anterior remains relatively relaxed. The increased swing phase EMG amplitude of the rectus femoris during backward running may indicate hip flexion before touchdown, contributing to the softer touchdown reported [8]. Our findings of distinctly different muscle activation patterns of selected muscles support the notion that backward running has the potential to train selective muscles in different ways forward running does. Limitations of this research are acknowledged. Although backward running speed was reduced compared to forward running speed similar metabolic costs could not be guaranteed
for each participant during the running tasks as individual backward running coordination levels of the novice backward runners may have differed. As our runners and triathletes did not have specific experience in backward running, results cannot be generalized to team sport athletes, who are exposed to backward running due to the multi-directional nature of their sports, or to athletes with solid backward running practice [3]. It remains to be seen whether backward running exercise results in more efficient muscle activity patterns and lesser increase in regional plantar loading due to altered coordination skills of athletes. Future research should investigate whether athletes having acquired automated backward running styles evoke modified plantar pressure patterns due to more economical coordination [3,30]. 5. Conclusion Backward running exhibits specific foot roll-over characteristics initiating at the forefoot during touchdown, progressing to the midfoot/rearfoot during midstance, and then progressing back to the forefoot to generate the push-off. This forefoot dominant locomotion style is characterized by higher loading of the forefoot compared to the normal heel–toe forward running style. Muscle activity was increased for backward compared to forward running, predominantly during stance and foremost at the thigh and hip muscles, although running speed was adjusted to account for higher metabolic costs during backward running. Future research should investigate whether training programmes for backward running decrease required muscle activity and reduce the high forefoot loading. Acknowledgements The authors thank Helga Raape for assistance during EMG measurements of MViC and Dieter Klein for technical support of this research. Conflict of interest statement The authors have no financial and personal relationships with other people or organizations that could inappropriately influence their work. References [1] Bates BT, Morrison E, Hamill J. Differences between forward and backward running. In: Adrian M, Deutsch H, editors. Proceedings 1984 Olympic Scientific Congress: Biomechanics, July 19–26. Eugene, OR, USA: University of Oregon Microform Publications; 1984. p. 127–35. [2] Threlkeld AJ, Horn TS, Wojtowicz GM, Rooney JG, Shapiro R. Kinematics, ground reaction force, and muscle balance produced by backward running. J. Orthop. Sports Phys. Ther. 1989;11:56–63. [3] Schubert C, Brauner T, Eckardt D, Sterzing T. Trainierbarkeit des Ru¨ckwa¨rtslaufens. 8. Gemeinsames Symposium der dvs-Sektionen Biomechanik, Sportmotorik und Trainingswissenschaft, Hamburg, Deutschland. In: Mattes, Wollesen, editors. dvs Band 204: Bewegung und Leistung—Sport, Gesundheit & Alter. Hamburg, Deutschland: Feldhaus Verlag Edition Czwalina; 2010. p. 44. [4] Nadeau S, Amblard B, Mesure S, Bourbonnais D. Head and trunk stabilization strategies during forward and backward walking in healthy adults. Gait Posture 2003;181:134–42. [5] Rosenbaum D, Schubert C, Sterzing T. Gait speed and plantar pressure pattern differences in forward and backward walking with eyes open and closed. In: Van Sint JS, et al., editors. Proceedings of the 23. Congress of the International Society of Biomechanics. Belgium: Vrije Universiteit Brussel; 2011. p. 190–1. [6] Jansen K, De Groote F, Massaad F, Meyns P, Duysens J, Jonkers I. Similar muscles contribute to horizontal and vertical acceleration of center of mass in forward and backward walking: Implications for neural control. J. Neurophysiol. 2012;107:3385–96. [7] DeVita P, Stribling J. Lower extremity joint kinetics and energetics during backward running. Med. Sci. Sports Exerc. 1991;23:602–10. [8] Cavagna GA, Legramandi MA, La Torre A. Running backwards: soft landing— hard takeoff, a less efficient rebound. Proc. Biol. Sci. 2011;278:339–46.
T. Sterzing et al. / Gait & Posture 46 (2016) 135–141 [9] Larson P, Higgins E, Kaminski J, Decker T, Preble J, Lyons D, McIntyre K, Normile A. Foot strike patterns of recreational and sub-elite runners in a long-distance road race. J. Sports Sci. 2011;29:1665–73. [10] Breine B, Malcolm P, Frederick EC, de Clercq D. Relationship between running speed and initial foot contact patterns. Med. Sci. Sports Exerc. 2014;46: 1595–603. [11] Cavagna GA, Legramandi MA, La Torre A. An analysis of the rebound of the body in backward human running. J. Exp. Biol. 2012;215:75–84. [12] Flynn TW, Soutas-Little RW. Mechanical power and muscle action during forward and backward running. J. Orthop. Sports Phys. Ther. 1993;17:108–12. [13] Chaloupka EC, Kang J, Mastrangelo MA, Donnelly MC. Cardiorespiratory and metabolic responses during forward and backward walking. J. Orthop. Sports Phys. Ther. 1997;25:302–6. [14] Terblanche E, Cloete WA, du Plessis PAL, Sadie JN, Strauss A, Unger M. The metabolic transition speed between backward walking and running. Eur. J. Appl. Physiol. 2003;90:520–5. [15] Hreljac A, Imamura R, Escamilla RF, Casebolt J, Sision M. Preferred and energetically optimal transition speeds during backward human locomotion. J. Sports Sci. Med. 2005;4:446–54. [16] Minetti AE, Ardigo` LP. The transmission efficiency of backward walking at different gradients. Pflugers Arch. 2001;442:542–6. [17] Flynn TW, Connery SM, Smutok MA, Zeballos RJ, Weisman IM. Comparison of cardiopulmonary responses to forward and backward walking and running. Med. Sci. Sports Exerc. 1994;26:89–94. [18] Williford HN, Olson MS, Gauger S, Duey WJ, Blessing DL. Cardiovascular and metabolic costs of forward, backward, and lateral motion. Med. Sci. Sports Exerc. 1998;30:1419–23. [19] Wright S, Weyand PG. The application of ground forces explains the energetic cost of running backward and forward. J. Exp. Biol. 2001;204:1805–15.
141
[20] Zampeli F, Moraiti CO, Xergia S, Tsiaras VA, Stergiou N, Georgoulis AD. Strideto-stride variability is altered during backward walking in anterior cruciate ligament deficient patients. Clin. Biomech. 2010;25:1037–41. [21] Heath EM, Blackwell JR, Baker UC, Smith DR, Kornatz KW. Backward walking practice decreases oxygen uptake, heart rate and ratings of perceived exertion. Phys. Ther. Sport 2001;2:171–7. [22] Hooper TL, Dunn DM, Erick Props J, Bruce BA, Sawyer SF, Daniel JA. The effects of graded forward and backward walking on heart rate and oxygen consumption. J. Orthop. Sports Phys. Ther. 2004;34:65–71. [23] Roos PE, Barton N, van Deursen RWM. Patellofemoral joint compression forces in backward and forward running. J. Biomech. 2012;45:1656–60. [24] Hughes J. The clinical use of pedobarography. Acta Orthop. Belg. 1993;59:10–6. [25] Rosenbaum D, Becker HP. Plantar pressure distribution measurements. Technical background and clinical evaluation. Foot Ankle Surg. 1997;3:1–14. [26] Sterzing T, Custoza G, Ding R, Cheung JT-M. Segmented midsole hardness in the midfoot to forefoot region of running shoes alters subjective perception and biomechanics during heel–toe running revealing potential to enhance footwear. Footwear Sci. 2015;7:63–79. [27] Hermens HJ, Freriks B, Disselhorst-Klug C, Rau G. Development of recommendations for SEMG sensors and sensor placement procedures. J. Electromyogr. Kinesiol. 2000;10:361–74. [28] Weist R, Eils E, Rosenbaum D. The influence of muscle fatigue on electromyogram and plantar pressure patterns as an explanation for the incidence of metatarsal stress fractures. Am. J. Sports Med. 2004;32:1893–8. [29] Cavanagh PR, Ulbrecht JS. Clinical plantar pressure measurement in diabetes: rationale and methodology. Foot 1994;4:123–35. [30] Mehdizadeh S, Arshi AR, Davids K. Quantifying coordination and coordination variability in backward versus forward running: implications for control of motion. Gait Posture 2015;42:172–7.