G Model GAIPOS 4211 No. of Pages 5

Gait & Posture xxx (2014) xxx–xxx

Contents lists available at ScienceDirect

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

Effects of footwear on treadmill running biomechanics in preadolescent children Karsten Hollander a,b , Dieko Riebe c, Sebastian Campe d, Klaus-Michael Braumann a , Astrid Zech c, * a

Institute of Sports and Exercise Medicine, University of Hamburg, Hamburg, Germany Orthopaedic Department, The Royal Children's Hospital, Melbourne, Victoria, Australia c Institute of Human Movement Science, University of Hamburg, Hamburg, Germany d Department of Sports Science, Otto-von-Guericke-University, Magdeburg, Germany b

A R T I C L E I N F O

A B S T R A C T

Article history: Received 25 November 2013 Received in revised form 9 May 2014 Accepted 12 May 2014

While recent research debates the topic of natural running in adolescents and adults, little is known about the influence of footwear on running patterns in children. The purpose of this study was to compare shod and barefoot running gait biomechanics in preadolescent children. Kinematic and ground reaction force data of 36 normally developed children aged 6–9 years were collected during running on an instrumented treadmill. Running conditions were randomized for each child in order to compare barefoot running with two different shod conditions: a cushioned and a minimalistic running shoe. Primary outcome was the ankle angle at foot strike. Secondary outcomes were knee angle, maximum and impact ground reaction forces, presence of rear-foot strike, step width, step length and cadence. Ankle angle at foot strike differed with statistical significance (p < 0.001) between conditions. Running barefoot reduced the ankle angle at foot strike by 5.97 [95% CI, 4.19; 7.75] for 8 km h 1 and 6.18 [95% CI, 4.38; 7.97] for 10 km h 1 compared to the cushioned shoe condition. Compared to the minimalistic shoe condition, running barefoot reduced the angle by 1.94 [95% CI, 0.19 ; 3.69 ] for 8 km h 1 and 1.38 [95% CI, 3.14 ; 0.39 ] for 10 km h 1. Additionally, using footwear significantly increased maximum and impact ground reaction forces, step length, step width and rate of rear-foot strike. In conclusion, preadolescent running biomechanics are influenced by footwear, especially by cushioned running shoes. Health professionals and parents should keep this in mind when considering footwear for children. ã 2014 Elsevier B.V. All rights reserved.

Keywords: Barefoot running Children Footstrike Footwear Running biomechanics

1. Introduction Gait and running patterns are influenced by numerous external and internal factors [1] and much evidence exists that footwear changes running kinetics and kinematics [2,3]. Adolescent and adult runners who wear cushioned shoes tend to land in a more dorsiflexed ankle angle and use predominantly a rear-foot strike (RFS) for initial ground contact [4]. Lieberman et al. [2] also show varying ground reaction forces (GRF) for cushioned shoes compared to barefoot running. In RFS, the collision of the heel with the ground generates an immediate large impact transient. In contrast, a forefoot strike, characterized by a more plantarflexed

* Corresponding author at: Institute of Human Movement Science, University of Hamburg, Mollerstrasse 2, 20148 Hamburg, Germany. Tel.: +49 40 42838 9310; fax: +49 40 42838 6268. E-mail address: [email protected] (A. Zech).

ankle angle, generates a smaller impact force without an impact transient. This leads to the presumption that habitually rear-foot striking runners have higher rate of injuries compared to those who mostly forefoot strike [5]. Whereas previous research majorly compared shod and barefoot conditions with a view on the running gait of adults or on the walking gait of children [6,7], little is known on the effects of footwear on running biomechanics in children[8]. Since children are not influenced by years of using cushioned shoes, one might speculate that running style and biomechanics are less affected by the use of footwear. Although this implies that children’s shod and barefoot running patterns are different to those of adults [9], to our knowledge, no study has yet compared running kinematics and kinetics between various footwear conditions in children. Consequently, the purpose of this study was to analyze the influence of footwear with no or different cushioning on running kinetic and kinematic parameters in 6–9 years old children. Based on previous studies in the adult population, we hypothesized that

http://dx.doi.org/10.1016/j.gaitpost.2014.05.006 0966-6362/ ã 2014 Elsevier B.V. All rights reserved.

Please cite this article in press as: Hollander K, et al.. Effects of footwear on treadmill running biomechanics in preadolescent children. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.05.006

G Model GAIPOS 4211 No. of Pages 5

2

K. Hollander et al. / Gait & Posture xxx (2014) xxx–xxx

running shod alters the ankle angle and foot strike pattern during initial ground contact of the running gait cycle. In order to differentiate between footwear with and without cushioning, we compared the barefoot condition with a traditional running shoe as well as a non-cushioned minimalistic shoe. 2. Methods The randomized crossover trial was performed in a University Biomechanics Laboratory. Thirty-six normally developed children were recruited from local sports teams. The participants had a mean  SD (range) age of 7.42  1.05 (6–9) years, height of 1.29  0.07 (1.14–1.43) m, weight of 26.42  3.71 (18.6–33.5) kg and a physical activity of 4.8  1.6 (2–10) h per week. Females comprised 61% (n = 22) of the study sample. Written informed parental consent and child assent was obtained. Parents were also allowed to be present during the test procedure. Ethical approval for the study was obtained from the local ethics committee. For inclusion, children had to be physically active and between 6 and 9 years, with a height of more than 1.10 and less than 1.45 m. Exclusion criteria were: history of orthopedic, musculoskeletal or neurological abnormalities likely to influence their gait. Participants were also excluded if they could not show comfortable gait patterns on the treadmill. Participating children and their parents were not informed about the study hypothesis. Demographic data and self-reported volume of sports activity (hours) were obtained by the main study investigator. The children were guided and instructed to practice running on the treadmill until they became comfortable with the test. Three different conditions were applied in random counterbalanced order: barefoot running (B), neutral cushioned running shoe running (C) and minimalistic running shoe running (M). The cushioned running shoe (Asics Galaxy1) had an ethylene-vinyl acetate midsole, a 8.4 mm heel-forefoot offset and a mass of 171 g (EUsize 33), while the minimalistic shoe (Leguanito1) had a polyvinyl chloride (LIFOLIT1) midsole, a 0 mm heel-forefoot offset and a mass of 145 g (EU-size 33). Primary outcome was ankle angle at foot strike. Secondary outcomes were knee angle at foot strike, maximum ground reaction force (GRF), impact GRF, rate of rearfoot strike (RFS), step width, step length and cadence. Running kinematics were determined using a three-dimensional 8-camera infrared motion analysis system operating at 200 Hz (VICON, Oxford Metrics, UK). Sixteen retro-reflective markers (14 mm diameter) were placed on anatomical landmarks of the pelvis and both lower extremities according to the Plug-inGait model (VICON, Oxford Metrics, UK). The anatomical landmarks included the anterior and posterior superior iliac spine, lateral thigh, lateral femoral epicondyle, lateral tibia, lateral malleolus, second metatarsal head and the most distal part of the calcaneus. All marker placements remained unchanged between running conditions except for the foot markers. For the shod conditions the posterior calcaneus and the second metatarsal head markers were applied directly on the shoes in reference to the foot. In order to ensure the comparability of marker positions between the shod and barefoot conditions, we used a two-fold strategy. The more superior position of the second metatarsal head marker during the running shoe condition was compensated by adjusting the superior–inferior position of the calcaneal marker. Both markers had similar distance to the ground. Separate calibrations were used for each shod and the barefoot condition. For data collection, the children ran at two different velocities (v1 = 8 km h 1 and v2 = 10 km h 1) with each of the three footwear conditions. Participants were given a one-minute rest between trials. The identical procedure was used for the next two footwear conditions. Data recording and processing were performed with Vicon Nexus 1.7.1

(VICON, Oxford Metrics, UK) and Matlab software (Mathworks, USA). The Plug-in-Gait dynamic lower body model was used to identify the thigh (aligned with the head of the femur and lateral femoral epicondyle), shank (aligned with the lateral femoral epicondyle and lateral malleolus) and foot (aligned with the calcaneus and metatarsal head) segments. Data has been filtered using a Woltring filtering routine (mean square error = 15). Initial foot-ground contact (foot strike) was defined when the vertical velocity of the distal heel marker changed from negative to positive [10]. A minimum of 50 gait cycles were analyzed (25 per leg) and trials were normalized to 100% of gait cycle. For statistical analysis the mean of the last 10% of the gait cycle was used. The Vicon motion capture system is a reliable tool for the analysis of gait kinematics [11] and was also used with children in other studies [7]. For kinetics, we used an instrumented treadmill (HP cosmos, Germany) with a capacitance-based pressure platform under the surface (FDM-THQ, Zebris Medical GmbH, Germany) [12]. The surface was 1.74  0.65 m containing a sensing area of 1.36  0.54 m and 10,240 sensors, with the size of 0.085  0.085 m each. Data were sampled at 300 Hz and the slope maintained in a horizontal position. The system captured the relative arrangement, the geometry and the applied pressure of each footfall as a function of time. The GRF-time curves were normalized to 100% of the stance phase. Matlab (Mathworks, USA) was used to calculate the maximum GRF during the propulsive period and to decide whether the rear-foot was used for initial ground contact. According to recent research[2], a rear-foot strike was identified by using the presence of an additional peak in the GRF curve (impact transient) during the contact period (first third of each running gait cycle). In case of an impact transient, its peak was calculated and defined as impact GRF. Maximum and impact GRF were normalized to body mass (% of body weight). Zebris software (Zebris Medical GmbH, Germany) was used to process the data into footfall patterns and to compute the temporal and spatial parameters cadence (steps per minute), step length, and step width for each foot. The step length is defined as the anterior– posterior and step width as the medial–lateral distance between the initial contacts of two steps (left and right foot). Cadence was calculated by using the number of steps during a 60 s interval. Mixed models were used to determine footwear differences in dependent variables (ankle and knee angle, maximum GRF, impact GRF, step width, step length and cadence). Children were included as a random factor. The interesting main effect of the shoe condition was included as a fixed effect and all models were adjusted for age, sex, height, weight and leg length. Tentatively, two-way interactions (shoe  sex and shoe  side (left/right)) were added and kept in all models if significant. In case of a significant effect, a Bonferroni post hoc test was conducted in order to identify differences across the three shoe conditions. The level of significance was defined as a = 0.05. For the dichotomous variable “presence of RFS" a generalized estimating equation model for a repeated measures logistic regression was calculated and for shoe comparisons odds ratios are presented. SPSS statistical package version 21 (SPSS, Armonk, NY, USA) was used for all statistical procedures. 3. Results All participants (n = 36) successfully completed the treadmill running protocol. 34 children accomplished the running tests at both velocities whereas two participants were only measured during the 8 km h 1 run due to difficulties with the speed at 10 km h 1. Spatiotemporal data of three participants could not be processed to statistical analysis due to missing data. A total of 210 running trials were registered and analyzed. Typical

Please cite this article in press as: Hollander K, et al.. Effects of footwear on treadmill running biomechanics in preadolescent children. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.05.006

G Model GAIPOS 4211 No. of Pages 5

K. Hollander et al. / Gait & Posture xxx (2014) xxx–xxx

3

Fig. 1. Ankle joint kinematics (A = cushioning shoe; B = barefoot; C = minimalistic shoe) and vertical ground reaction forces (D = cushioning shoe; E = barefoot; F = minimalistic shoe) for a typical subject at 8 km h 1 during the stance phase of the gait cycle. Bold lines represent the mean curve and dashed lines the 95% confidence interval of the mean difference between curves of multiple gait cycles in a time series. In this case, no impact transient is seen in the barefoot and minimalistic shoe condition.

Fig. 2. Mean ankle joint kinematics for all children running at 8 km h 1. Ankle angles differed statistical significantly between all conditions in the initial phase of ground contact (last 10% of gait cycle).

representations of ankle joint kinematics and vertical ground reaction forces related during the stance phase of barefoot, minimalistic and cushioned shoe running are shown in Fig. 1. Mean gait running cycle ankle angles of all children differed between the three footwear conditions (Fig. 2). Ankle angles at initial ground contact (90–100% of each running gait cycle) for both velocities were statistically significantly different (p < .001) between footwear conditions. The dorsiflexion was smallest during barefoot running and greatest during cushioned shoe conditions (Table 1). All pairwise comparisons (except barefoot vs. minimalistic shoe at 10 km h 1) were statistically significant (p < .05) (Tables 2 and 3). No differences between footwear conditions were found for the knee angle at foot strike. For both tested velocities, children have the shortest step length and highest cadence in barefoot running whereas cushioned shoe running produced longest step length and lowest cadence (Table 1). All post hoc pairwise comparisons of both outcomes were significantly different (p < .001). The mean rate of RFS (8 km h 1/10 km h 1) was highest during the cushioned shoe running (75.8%/78.5%), followed by minimalistic shoe running (63.6%/67.7%) (Tables 2 and 3). The smallest rate of RFS occurred in barefoot running (54.5%/60.0%). Comparisons

Table 1 Group mean (SD) for continuous data of spatio-temporal, kinematic and kinetic parameters and total rates for dichotomous data (analyzed at 8 and 10 km h Cushioned shoe 8km h Ankle amplitude ( ) Knee amplitude ( ) Maximum GRF (% of body weight) Impact GRF (% of body weight) Step width (cm) Step length (cm) Cadence (steps per minute) Total rates of RFS (%)

1

7.9 (6.9) 11.8 (5.7) 216.2 (20.2) 103.6 (66.1) 5.9 (1.8) 71.7 (5.9) 186.6 (15.6) 75.8

Barefoot 10km h

1

8.0 (7.1) 12.9 (5.5) 222.0 (19.9) 134.5 (68.1) 5.7 (1.6) 86.6 (7.1) 193.1 (16.4) 78.5

8km h

1

2.0 (5.4) 11.5 (4.8) 210.2 (19.6) 64.7 (61.7) 5.6 (1.7) 68.6 (5.3) 196.1 (15.3) 54.5

1

)

Minimalistic shoe 10km h

1

1.9 (6.3) 12.2 (4.8) 215.1 (21.7) 88.8 (72.4) 5.3 (1.5) 81.4 (7.4) 205.9 (18.6) 60.0

8km h

1

3.9 (5.7) 10.9 (6.4) 216.0 (21.1) 87.2 (65.9) 5.6 (2.0) 69.3 (5.5) 192.8 (14.7) 63.6

10km h 3.2 12.2 220.2 108.0 5.5 83.4 200.6 67.7

1

(6.4) (6.1) (22.0) (73.1) (1.7) (7.0) (16.8)

SD: Standard deviation; GRF: ground reaction force (during mean of all running gait cycle); RFS: rear-foot strike.

Please cite this article in press as: Hollander K, et al.. Effects of footwear on treadmill running biomechanics in preadolescent children. Gait Posture (2014), http://dx.doi.org/10.1016/j.gaitpost.2014.05.006

G Model GAIPOS 4211 No. of Pages 5

4

K. Hollander et al. / Gait & Posture xxx (2014) xxx–xxx

Table 2 Mixed model effects [95% CI] of pairwise comparisons between conditions for 8 km h

Ankle amplitude Knee amplitude Maximum GRF Impact GRF Step width Step length Cadence Total rates of RFS

and odds ratio [95% CI] for “total rates of RFS”.

Tests on univariate

Pairwise comparisons (effect [CI])

C  B  M (p-value)

CB

p-value

CM

p-value

BM

p-value

0.001 0.488 0.001 0.001 0.098 0.001 0.001 NA

5.97 [4.19; 7.75] 0.17 [ 1.52; 1.79] 6.03 [1.77; 10.30] 21.09 [13.31; 28.86] 0.28 [ 0.11; 0.67] 3.04 [2.08; 4.00] 9.45 [ 13.03; 5.87] 2.72 [1.25; 5.94]

Effects of footwear on treadmill running biomechanics in preadolescent children.

While recent research debates the topic of natural running in adolescents and adults, little is known about the influence of footwear on running patte...
578KB Sizes 3 Downloads 4 Views