Shoes Influence Lower Limb Muscle Activity and May Predispose the Wearer to Lateral Ankle Ligament Injury

Transcription

1 Shoes Influence Lower Limb Muscle Activity and May Predispose the Wearer to Lateral Ankle Ligament Injury Robin Kerr, Graham P. Arnold, Tim S. Drew, Lynda A. Cochrane, Rami J. Abboud Institution of Motion Analysis & Research, Orthopaedic & Trauma Surgery, TORT Centre, Ninewells Hospital & Medical School, University of Dundee, Dundee DD1 9SY, Scotland, United Kingdom Received 4 June 2007; accepted 16 June 2008 Published online 8 October 2008 in Wiley InterScience ( DOI /jor ABSTRACT: Lateral ankle ligaments are injured by hyperinversion of the foot. Foot position is controlled by the lower limb muscles. Awareness of foot position is impaired by wearing shoes. We aimed to determine the influence of wearing shoes upon muscle activity. Sixtytwo healthy subjects underwent the same measurements, barefoot and with standardized shoes in a random order. Electromyography (EMG) was recorded from the peroneus longus muscle in response to sudden and unanticipated inversion of the ipsilateral foot. Following foot inversion, the EMG signal showed an initial peak muscle contraction followed by a sustained smaller contraction. Both changes were significantly greater in shoes compared to the barefoot condition for all tested degrees of inversion. Muscle contraction following sudden inversion of the foot was significantly greater when wearing shoes. This greater muscular contraction may be an intrinsic mechanism to oppose the increased moment created by the inverted foot/shoe condition, and hence, may counter balance the increased tendency to injure the lateral ankle ligaments created by wearing shoes. ß 2008 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 27: , 2009 Keywords: ankle; footwear; injury; electromyography Correspondence to: Rami J. Abboud (T: þþ(0) ; F: þþ(0) ; r.j.abboud@dundee.ac.uk) ß 2008 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. The lateral ligaments of the ankle are the most frequently injured structures in an athlete s body. 1 To reduce injury incidence, the mechanism and contributing factors must be understood. Injury occurs predominantly when the ligaments are stressed in forced hyperinversion and plantarflexion, 1 4 with adduction of the foot also playing a role. 1,4 Thus, lateral ligament injuries are due to supination of the weightbearing foot. Injuries occur when the external inversion moment (the reaction force acting at the ground contact point relative to the center of rotation of the ankle and subtalar joints) is larger than the internal eversion moment created by the lateral ligaments and peroneal muscles. 5 Wearing shoes increases the inverting leverage because the sole raises the foot s plantar surface from the ground. Thus, shoes increase the tendency of the foot to rotate about the ankle and subtalar axes. We hypothesized that shoes predispose the wearer to inversion lateral ligament injury, and consequently influence the protective everting lower limb muscle response. Although inversion of the foot coinciding with loading the ankle results in injury (e.g., when landing from a jump), tripping during the gait cycle, causing inversion of the foot prior to ankle loading, will allow time for a corrective everting response from the peroneal muscles. 2,6,7 A clear understanding of the relationship between foot position in inversion and the origins of a correcting neuromuscular response from the mechanoreceptors of the skin, ankle ligaments, joint capsule, lower limb muscles, or tendons has not emerged However, it seems logical that impairment of afferent proprioceptive capabilities would affect the efferent muscular response. Robbins et al. 11 found that foot position awareness, and therefore proprioception, was significantly impaired by wearing shoes. They concluded that this was due to loss of plantar tactile sensitivity. The shoe sole provides an interface between the highly innervated sensory surface of the plantar sole of the foot and the ground contour that the body interprets. If wearing shoes diminishes sensory information obtained from the proprioceptive sensory receptors, then postural corrections in response to changes in ground contour may be affected, further supporting our hypothesis that wearing shoes predisposes the lateral ankle ligaments to injury. A change in ground contour produces inversion of the foot, with the response being activity of the lower limb muscles responsible for positioning the foot. A sudden change in ground contour can be created by a moving floor platform, with lower limb muscle activity measured using electromyography (EMG). 3,6,12 15 Although injury to the lateral ligaments occurs during gait due to weight bearing on a supinated foot, 1 4 the most realistic way to replicate this statically is with sudden inversion, because inversion predominantly stresses the lateral ligaments. 16 Previous studies of neuromuscular response to sudden foot inversion compared EMG signals obtained from groups with healthy and previously injured ankles. One study reported significant differences between such groups for peroneus longus reaction time, 14 whereas others found no differences. 3,12,13,15 Other studies assessed the effect of sudden inversion on the lower limb muscles responsible for positioning the inverted foot, but none have assessed the influence of wearing shoes. We proposed that shoes predispose the wearer to inversion injury of the lateral ligaments, and in turn, affect the protective everting response of the peroneal muscles. 318

2 SHOES MAY PREDISPOSE THE WEARER TO ANKLE LIGAMENT INJURY 319 METHODS AND MATERIALS Subjects Following approval by the local Research and Ethics Committee, 62 subjects (41 males, 21 females), consented to participate. Subjects were between 18 and 55 years old without current or previous ankle injury. Instrumentation The apparatus consisted of three synchronized sections: a personal computer, a portable EMG system, 17,18 and a computer-controlled two-footplate tilting platform 19 (Fig. 1). A custom multitasking program was developed to provide individual control of the footplate platforms and record EMG signals. The program allowed footplate movement sequences to be customized, saved, and repeated, while a timing facility synchronized the EMG data recording to the foot plate movement sequences at a sampling rate of 200 Hz. The rate was enforced due to the limitation in the synchronization protocol with the footplate system. The highest EMG signal frequency of the lower limb muscles is 116 Hz, 20 but further investigation (to confirm that no aliasing effect took place) was conducted by analysing the power frequency spectrum of the Peroneus Longus signal using a rate of 2000 Hz from 16 subjects. The highest 95% power frequency was found to be equal to 87 Hz. The computer collected EMG data and sent information to the two-footplate tilting platform, which rotated at an angular velocity of 1008/s to control inversion. Recordings were made of the angle of each footplate in the frontal plane relative to the test protocol starting position (with the footplates parallel to the laboratory floor) and of the EMG signal. Footplate Inversion Footplate inversion was limited to 208 to remain within the safe range of foot/shoe motion 15 and avoid injury. The protocol included footplate inversion measures at 0, 10, and 208. The incremental steps of and were investigated to determine significant differences that might provide an understanding of the peroneus longus response to sudden and unanticipated inversion of the ipsilateral foot. The apparatus incorporated three safety features: the upper footplate surfaces were covered with adhesive material to decrease the susceptibility of the feet to slip; reinforcement blocks positioned beneath the footplates, limited their range of motion should the software fail to limit inversion to 208; and two parallel, vertical handrails were fitted to the wall anterior to the footplates to reassure the subjects. Pre-experimental Protocol Surface EMG silver/silver chloride electrodes were attached to the peroneus longus muscle of each lower limb. 17,18 One electrode was located superficially to the bony insertion of the muscle; the second was positioned superficially to the motor end point (MEP) of the muscle. The MEP was located by stimulating the relevant muscle with a single electrical pulse. Electrical stimulation would elicit a muscular contraction only if applied to the MEP proximity. The skin at the six electrode sites was abraded to remove dead cells and oils, which aid in reducing electrical impedance. 17,18,21 Electrode gel served as a conductive medium. Two reference electrodes were positioned superficially to each of the subject s clavicles. The subject then completed a sequence of dorsiflexion and eversion movements to ensure that the electrodes were properly connected and that satisfactory EMG signals were being obtained. Experimental Protocol One standardized model of neutral commercial leisure trainers (Donnay 1 Intl, DM 04 00R, Livingston, UK) was used. The researcher tied the shoe laces to ensure consistent fit. When stepping onto the footplates, subjects were asked to stand with each foot in the center of each footplate, which was in line with the axis of rotation of the footplate, with their weight distributed evenly between each lower limb. Subjects held the handrails very loosely and faced the wall during data collection to ensure that changes in body posture brought about by the footplates were primarily corrected by the lower limb muscles and not the upper body. Each subject acted as their own control by completing the same footplate inversion movements barefoot and with shoes. Footplate inversion movements were conducted in two repetitive cycles of the same limb (Fig. 2). Data were recorded during these two sequences, left inversion and right inversion. Subjects completed a two other sequences of different footplate movements from which data were not recorded. This prevented subjects from becoming familiar with the movements from which data were recorded. Each test sequence was completed twice; each footplate movement in a sequence was maintained for 5 s. The study sample was divided randomly into four groups: half undertook the test sequences first barefoot then with shoes, with one-quarter starting barefoot with left inversion and one-quarter starting barefoot with right inversion. The other half undertook the test sequences first with shoes then barefoot, with a quarter starting with left inversion and a quarter with right inversion. Figure 1. Two-footplate tilting platform. Data Processing The EMG signals were rectified. To test the hypothesis that shoes influence lower muscle activity, a range of measures of muscle activity was assessed. Ten variables were extracted for each test sequence using custom designed software (Table 1 and Fig. 3). Reaction time was defined as the time elapsed from the beginning of the footplate trigger to the time of maximum rectified EMG signal in the 0.3 s following the footplate trigger.

3 320 KERR ET AL. Figure 2. EMG signal from left peroneus longus barefoot and with shoes (each showing a test sequence, i.e., two repetitive cycles). The average amplitude was calculated as the mean of the EMG signal values over the final 2.5 s in which the subject s foot and the footplate were held in the inverted position. This allowed evaluation of the EMG signal after it had settled from its initial peak contraction due to sudden footplate inversion. Statistical Methods Repeated-measures ANOVA was conducted with within-subjects factors of side (right, left), footplate inversion (0 108,10 208, 0 208), foot condition (barefoot, shoes), cycle (1, 2), and test sequence (1, 2). The Shapiro-Wilks procedure was used to test for normality. The assumption of sphericity was examined with Mauchly s test, and the Huynh-Feldt correction was applied where appropriate. Post-hoc comparisons were made using Bonferroni s correction. The order in which subjects were exposed to footplate inversion was determined using a Latin squares balanced design to avoid confounding this factor with order. Table 1. Definitions, Abbreviations, and Units for Table 2 Variable Abbreviation Units Reaction time moving from of inversion rt0.10 ms Reaction time moving from of inversion rt0.20 ms Reaction time moving from of inversion rt10.20 ms Peak of EMG signal moving from of inversion peak0.10 V Peak of EMG signal moving from of inversion peak0.20 V Peak of EMG signal moving from of inversion peak10.20 V Average amplitude of EMG signal in starting position av0 V Average amplitude of EMG signal after moving from of inversion av0.10 V Average amplitude of EMG signal after moving from of inversion av0.20 V Average amplitude of EMG signal after moving from of inversion av10.20 V

4 SHOES MAY PREDISPOSE THE WEARER TO ANKLE LIGAMENT INJURY 321 sustained contraction of constant phase for the remainder of the 5 s during which the subject s foot was held at that inverted position. Peak and sustained contractions were consistently greater when subjects wore shoes compared to barefoot (Fig. 2). No significant differences were found between cycles, test sequences, or sides, and no interactions of these factors with other main effects were found. Therefore, figures in Table 2 are marginal means (i.e., averaged over cycles, test sequences, and sides) comparing the 10 EMG test variables between the 62 subjects barefoot and with shoes. No significant differences occurred between the reaction times measured barefoot or with shoes. All of the peak EMG signals were significantly greater with shoes. No significant differences occurred between the average EMG signal amplitude while subjects stood with the footplates parallel to the laboratory floor barefoot or with shoes. The average amplitude when the foot was inverted was significantly greater with shoes for all degrees of inversion. No significant differences were found between reaction time, peak contraction, and average amplitude for each footplate inversion (0 108, ,0 208). Figure 3. Diagrammatic representation of variables. RESULTS Foot inversion corresponded with an initial peak in the EMG amplitude of the ipsilateral peroneus longus muscle. This peak contraction was followed by a DISCUSSION Our results demonstrate that wearing shoes influences lower limb muscle activity. Peroneus longus muscle activity in response to sudden foot inversion was significantly greater when wearing shoes compared to the same degree of inversion while barefoot under the same conditions. Lateral ankle ligaments injury occurs when the moment of external inversion exceeds the moment of internal eversion. 5 An increase in the moment of external inversion within the hindfoot occurs when wearing shoes compared to barefoot. 5 These same principles can be applied to the inverted foot standing upon a footplate assuming a simplified model of static equilibrium (Fig. 4). When wearing shoes, the lever arm (x s ) to the ground reaction force (F rs ) is increased compared to the same degree of inversion while barefoot (x s > x b ), so the moment of external inversion (F rs x s )is greater. To maintain rotational equilibrium and prevent injury while one foot is inverted upon the platform, the moments of external and internal eversion must be equal (F rs x s ¼ F ms y s ). As the everting muscle force in shoes (F ms ) has the same length of lever arm about the subtalar joint as when barefoot (y s ¼ y b ), the everting muscle force in shoes (F ms ) must increase to maintain equilibrium. This increased force of muscle contraction corresponds to the significant increase in the EMG signal amplitude 21 of the everting peroneus longus muscle while wearing shoes. Therefore, the increased everting muscle activity while wearing shoes may be an intrinsic mechanism to counter the increased tendency to invert the foot and injure the lateral ligaments created by wearing shoes. Ideally, the degree of inversion would be assessed at the peak in EMG amplitude, a deficiency that has been addressed in the current study.

5 322 KERR ET AL. Table 2. Comparison of Barefoot and Shoes Measures 95% Confidence Interval Variable a Effect Mean Standard Error Lower Upper Significance rt0.10 Barefoot Shoes rt0.20 Barefoot Shoes rt10.20 Barefoot Shoes peak0.10 Barefoot <0.001 Shoes peak0.20 Barefoot <0.001 Shoes peak10.20 Barefoot <0.001 Shoes av0 Barefoot Shoes av0.10 Barefoot Shoes av0.20 Barefoot <0.001 Shoes av10.20 Barefoot <0.001 Shoes a Definitions and units for variables given in Table 1. A number of assumptions were made in the free body diagrams (Fig. 4). The foot was assumed stationary within the shoe during inversion and the footplate angle was assumed to correspond to the degree of rotation of the subtalar joint, barefoot or with shoes. The ground reaction force (F rb or F rs ) is a point load and does not change position. However, its lever arm will be proportionally greater with shoes (x s ) than when barefoot (x b ), regardless of where on the foot the force acts. The component of the ground reaction force (F rb or F rs ) parallel to the footplate is opposed by the friction force that acts to maintain static equilibrium. Therefore, the foot does not slide on the platform. The ability of the subject to estimate foot position on an angulated surface is impaired by wearing shoes, 11,22 suggesting that wearing shoes impairs proprioception. 11 One way in which a patient with a dorsal root ganglionopathy compensated for the proprioceptive impairment of their lower limbs was to increase their background muscle activity to stiffen their posture, particularly with their torso muscles, to achieve stability. 23 The EMG signals for the patient with a complete proprioceptive loss in both lower limbs 23 are not fully analogous with our study. However, they suggest that when uncertain of position as a result of diminished proprioception, the subject increases muscle activity to stiffen and stabilize the body. In our study, we found that muscle activity was increased when wearing shoes. When standing on the footplates positioned parallel to the laboratory floor, muscle activity was increased, although not significantly, when wearing shoes. Following foot inversion, the activity of the everting peroneus longus muscle was significantly greater when wearing shoes, both in initial peak contraction response and as the average amplitude while held in the position. This increased activity might be an intrinsic mechanism to compensate for diminished proprioception 23 induced by wearing shoes. 11 We believe our study is the first to compare muscle reaction times following sudden foot inversion, barefoot and with shoes. Reaction time of the peroneus longus muscle to sudden inversion of the ankle and subtalar joints was unaffected by shoes. If wearing shoes impairs proprioception, the functioning of proprioceptive structures more proximal to those encased in the shoe were unaffected, in accordance with the findings of Konradsen et al. 10 No significant differences existed between measures of reaction time, peak contraction, and average amplitude for each footplate inversion (0 108,10 208,0 208), reinforcing that the differences in lower limb muscle activity were due to wearing shoes and not the degree of footplate movement. Therefore, future work can be conducted with a single inversion of Shoes increase the tendency of the foot to invert at initial contact, in side-stepping movements 5 and when running, 24 while decreasing foot positional awareness. 11,22 A proprioceptive impairment induced by wear-

6 SHOES MAY PREDISPOSE THE WEARER TO ANKLE LIGAMENT INJURY 323 limb. Injuries to the lateral ankle ligaments are due to weight bearing on a supinated foot, 1 4 a condition that the foot may be predisposed to by wearing shoes. This crucial component of swing occurs 220 ms before heel contact, 2,6 which may allow time for the lower limb muscles to correct the position of the supinated foot. Even if partially corrected, wearing shoes will increase the tendency of the foot to invert again at initial contact 24 when it is loaded with body weight, potentially leading to injury to the lateral ankle ligaments. Thus, it appears that wearing shoes predisposes the lateral ankle ligaments to injury. Walking barefoot is usually impractical. Therefore, future footwear developments should look to minimize any proprioceptive impairment created by the shoes and have smaller sole heights to reduce the moment that may predispose the foot to invert and cause lateral ankle ligament injury. We are evaluating a range of other shoe brands and models in relation to muscle activity. ACKNOWLEDGMENTS Partial funding for this research was provided by The Wishbone Trust. The authors have no professional or financial affiliations that could have biased the presentation. The authors wish to thank Mr. I. Christie for his illustrations and Mr. G. Boath for his technical assistance. Figure 4. Free body diagram of inverted left foot in frontal place, posterior view, barefoot and with shoes. ing shoes has been suggested, 11 as also suggested by the EMG signals obtained from our study. During the mid to late swing phase of gait, the lateral aspect of the foot, inverted at about 108, passes within 5.5 mm clearance of the ground surface at its maximum horizontal velocity. 2,25 This element of swing requires precise proprioception to avoid contact between the lateral border of the foot and the ground. 7 The inversion angle adopted by the foot during swing is set relative to the perceived inversion angle from the proceeding stance phase, 22 which is impaired by wearing shoes. 11,22 Therefore, wearing shoes could impair an individual s ability to position their foot accurately during swing, which is crucial in avoiding injury. 7 In a cadaveric study, a swing phase inversion angle of 208, instead of the normal 108, led to contact with the ground and maximal supination of the foot within the confines of the test apparatus. 2 At this stage in the gait cycle, the body s center of gravity is anterior to the toes of the contralateral limb in stance; 25 thus, contact between the ground and the swing limb will lead to weight bearing on the supinated foot of the swing REFERENCES 1. Garrick JG The frequency of injury, mechanism of injury, and epidemiology of ankle sprains. Am J Sports Med 5: Konradsen L, Voigt M Inversion injury biomechanics in functional ankle instability: a cadaver study of simulated gait. Scand J Med Sci Sports 12: Nawoczenski DA, Owen MG, Ecker ML, et al Objective evaluation of peroneal response to sudden inversion stress. J Orthop Sports Phys Ther 7: Safran MR, Benedeff RS, Bartolozzi R, et al Lateral ankle sprains: a comprehensive review. Part 1: etiology, pathoanatomy, histopathogenesis, and diagnosis. Med Sci Sports Exerc 31:S429 S Stacoff A, Steger J, Stüssi E, Reinschmidt C Lateral stability in sideways cutting movements. Med Sci Sports Exerc 28: Konradsen L, Voigt M, Hojsgaard C Ankle inversion injuries. The role of the dynamic defence mechanism. Am J Sports Med 25: Konradsen L Sensori-motor control of the uninjured and injured human ankle. J Electromyogr Kinesiol 12: Feuerbach JW, Grabiner MD, Koh TJ, et al Effect of ankle orthosis and ankle ligament anaesthesia on ankle joint proprioception. Am J Sports Med 22: Freeman MAR, Dean MRE, Hanham IWF The etiology and prevention of functional instability of the foot. J Bone Joint Surg 47B: Konradsen L, Ravn JB, Sorensen AI Proprioception at the ankle: the effect of anaesthetic blockade of ligament receptors. J Bone Joint Surg 75B: Robbins S, Waked E, McLaren J Proprioception and stability: foot position awareness as a function of age and footwear. Age Ageing 24:67 72.

7 324 KERR ET AL. 12. Ebig M, Lephart SM, Burdett RG, et al The effect of sudden inversion stress on EMG activity of the peroneal and tibilalis anterior muscles in the chronically unstable ankle. J Orthop Sports Phys Ther 26: Isakov E, Mizrahi J, Solzi P, et al Response of the peroneal muscles to sudden inversion of the ankle during standing. Int J Sports Biomech 2: Konradsen L, Ravn JB Ankle instability caused by prolonged peroneal reaction time. Acta Orthop Scand 61: Vaes P, Duquet W, Gheluwe BV Peroneal reaction times and eversion motor response in healthy and unstable ankles. J Ath Train 37: Riegger CL Anatomy of the ankle joint. Phys Ther 68: Abboud RJ, Rowley DI, Newton DI Lower limb dysfunction may contribute to foot ulceration in diabetic patients. Clin Biomech 15: Abboud RJ Temporal relationships between foot pressure and muscle activity during walking. PhD thesis, Scotland UK: University of Dundee. 19. Abboud RJ An investigation into a relationship between lower limb muscle activity and ankle/foot sensation. Final report to The Wishbone Trust. 20. Wakeling JM, Tscharner VV, Nigg BM, Stergiou P Muscle activity in the leg is tuned in response to ground reaction forces. J Appl Physiol 91: Basmajian JV, De Luca CJ Muscles alive. Their functions revealed by electromyography, 5th ed. Baltimore, MD: Williams and Wilkins. 22. Waddington G, Adams R Football boot insoles and sensitivity to extent of ankle inversion movement. Br J Sports Med 37: Bloem BR, Allum JHJ, Carpenter MG, et al Triggering of balance corrections and compensatory strategies in patients with total leg proprioceptive loss. Exp Brain Res 142: Stacoff A, Nigg BM, Reinschmidt C, et al Tibiocalcaneal kinematics of barefoot versus shod running. J Biomech 33: Winter DA Kinematics. In: Biomechanics and motor control of human gait: nrmal, elderly and pathological, 2nd ed. Waterloo: Waterloo Biomechanics. p