16th June 2015
Orthotic Heel Wedges do not alter Hindfoot Kinematics and Achilles Tendon Force during Level and Inclined Walking in Healthy Individuals
Robert A Weinert-Aplin,1,2 Anthony M J Bull,2 Alison H McGregor1
1 Department of Surgery and Cancer, Imperial College London, London, U.K.; 2 Department of Bioengineering, Imperial College London, London, U.K.
Funding: This study was funded by an EPSRC Case award, with financial contributions from Vicon Motion Systems for the running costs of the project.
Conflicts of interest statement: None
Correspondence Address:
Robert Weinert-Aplin,
Department of Bioengineering,
Royal School of Mines,
South Kensington Campus,
Imperial College London,
London, SW7 2AZ,
United Kingdom
Abstract
Conservative treatments such as in-shoe orthotic heel wedges to treat musculoskeletal pathologies are not new. However, the mechanical basis by which such orthoses act have not been elucidated. Quantifying the mechanical changes that occur when wearing heel wedges may help to explain the mixed evidence supporting their use in management of Achilles tendonitis.
A musculoskeletal modelling approach was used to quantify changes in lower limb mechanics when walking due to the introduction of 12mm orthotic heel wedges. A control group of 19 healthy volunteers walked on a level and inclined walkway while optical motion, forceplate and plantar pressure data were recorded as model inputs.
Heel wedges induced a posterior shift of centre of pressure that resulted in increased ankle dorsi-flexion moments and reduced plantar-flexion moments. Consequently, this resulted in increased peak ankle dorsi-flexor muscle forces during early stance and reduced Tibialis Posterior and toe flexor muscles forces during late stance. Heel wedges did not reduce triceps surae [RWA1]muscle forces during any walking condition.
These results add to the body of clinical evidence that does notagainst the use of support heel wedges hypothesised to reduce Achilles tendon loading, as a means to treat Achilles tendonitis according to the hypothesis that heel wedges reduce Achilles tendon load during walking [RWA2]. our and the findings provide an explanation as to why this may be the casetheory is not appropriate.[MAH3]
Keywords: tendonitis, tendinitis, musculoskeletal, modelling, conservative treatment
Word Count: 3427
Introduction
The Achilles tendon is functionally important as it is the main driver of ankle motion during locomotion[RWA4]. Consequently it is a highly loaded tendon, with reported loads of up to 5 times body weight (BW) during level walking1 and 8.2BW- 12.5BW during running.2,3 With such high cyclic loads being experienced by the Achilles, it is unsurprising that the Achilles is a common site of overuse injury,4 with 5-18% of all lower extremity injuries involving the Achilles tendon.5 As a result of this high injury prevalence, there have been a number of reviews into Achilles injuries.6-11
Risk factors for Achilles injuries include: magnitude of Achilles tendon load, inappropriate equipment such as inflexible shoes, low heel tabs on running shoesinappropriate footwear12[RWA5], training errors7 and abnormal kinematics.6 Abnormal kinematics are generally considered to be related to over-pronation of the subtalar joint hindfoot [RWA6]causing asymmetric loading across the Achilles tendon.7 While there is some evidence to show differences in strain across the tendon cross-section13 and along the length,14 over-pronation alone has not been linked to injury risk in running.5,15-17 It has also been shown that differences exist between Medial Gastrocnemius and Soleus contractile behaviour during walking,18,19 running20 and cycling,21 suggesting that differential strains across the tendon may exist naturally during these activities.
Treatments for tendinopathies have varied substantially over the years, but have always been aimed at reducing or eliminating pain in the tendon. Conservative treatments which directly address the Achilles tendon include: have included: Rest Ice Compression Elevation, eccentric strengthening exercises,22-25 insoles and splints24,26 and ultrasound.27 Of the treatments that influence mechanical alignment of the hindfoot, insoles and splints are the only option24,26, but have been shown
The use of insoles or orthotics to correct abnormal hindfoot motion and correct for biomechanical mal-alignments to reduce pain and aid in return to sport has been shown to be possiblefollowing Achilles tendinopathy[RWA7].17 However, the direct link between correcting hindfoot motion and tendon healing has not been established, but this treatment is still recommended.9
Currently it is believed that incorporating heel wedges aids in reducing tendon strain during activities to avoid excess tendon loading and reduce pain during running.26,28,29 Investigations regarding heel wedges which plantarflex the ankle by raising the heel equally on the medial and lateral sides[RWA8], insoles and other orthotics have had variable success regarding reduction of pain and ability to return to sport.6,20,26,28,30-33 However, as orthotics require no invasive procedures and injury management can be at home and on-going, they are particularly appealing. ThisA similar approach has been used in osteoarthritis, where knee adduction moment has been the target of a variety of orthotics that aimed to alter the mechanical loading of the knee by altering the frontal plane alignment of the hindfoot[RWA9], with several studies showing positive effects of using foot orthotics on knee adduction moment34 and tibio-femoral load.35 However, while studies investigating such orthotics have been focussed on influencing out of plane moments at the knee, their investigations of orthotics focussed on relieving Achilles tendon strain through plantarflexion of the ankle effect on the ankle is often not reportedare less common[RWA10]. With wedges being able to alter loading at the knee, it is not unreasonable to hypothesise that a similar approach could alter the loading at the ankle, and indeed, this is the basis by which heel wedges are thought to operate during walking when managing Achilles tendon pain in Achilles Tendonitis patients.36 Therefore, the hypothesis of this study was that heel wedges which raise the heel are able to reduce Achilles tendon force during level and inclined walking.[RWA11]
The aim of this study was to quantify the effect of heel wedges on lower limb mechanics, specifically ankle joint angles, moments and muscle forces; and relate these to common injuries such as Achilles tendonitis. This is conducted during level and inclined walking in order to understand their effect beyond the constrained context of level walking.
Methods
The subject participant group consisted of nineteen healthy individuals, with no history of ankle injuries and no lower limb injury in the last 12 months and no clinical symptoms of Achilles Tendinopathies (8 male [mean (SD); age: 28 (3); height: 1.76 m (0.10); mass: 73.4 kg (12.0)] and 11 females [age: 29 (6); height: 1.63 m (0.05); mass: 58.7 kg (10.2)]. Individuals were excluded if they had ever been diagnosed with Achilles Tendinopathy or had any previous musculoskeletal or neuromuscular condition of the lower limb. Ethical approval was obtained by a university institutional review board in accordance with the Declaration of Helsinki and all participants were given an information sheet and provided a signed consent form upon arrival.
A pair of commercially available orthotic heel wedges were used by all subjects participants (Elevator Proheel™, Talar Made Orthotics Ltd, Springwood House, Foxwood Way, Chesterfield, Derbyshire, England) (Figure 1). The wedges are made from medium density ethylene vinyl acetate (EVA) foam and are designed to mould to the rearfoot and elevate the heel, with the aim of being used as a tendonitis treatment. Wedges were available for UK shoe sizes 2-5; 6-9 and 9.5-12.5 and all wedges had a 12mm rise from the front edge of the wedge to approximately where the centre of the heel would be. Subjects Participants were fitted with wedges corresponding to their running shoe size. All participants wore standard running shoes that they felt comfortable in. Shoes were checked visually for excessive wear under the sole and participants confirmed that they had used their running shoes previously before participating in the study[RWA12]. A wedge height of 12mm was chosen[RWA13] as it represents a height that is recommended for Achilles tendonitis patients.36
The mechanical effect of heel wedges on a variety of walking conditions necessitated the need for an inclined walkway which could securely accommodate a forceplate either on a level or inclined surface, shown in (Figure 2[RWA14]). The setup allowed the subjects participants to approach the 10° incline on a level surface for several steps, before ascending the 2m inclined section where foot strike was recorded followed by a few steps of level walking at the top of the incline. Subjects Participants were then asked to turn around and walk down the incline to provide the data for the downhill walking condition. For both inclined and declined walking, all subjects participants required three steps to cover the 2m inclined section, with the middle step being used for subsequent analysis. Participants were given as long as they needed to familiarise themselves with each condition. This was assessed by participants themselves as they walked freely around the laboratory and up and down the level and inclined walkway untiltill they felt comfortable. The amount of time taken by each individual was not measured, but was on the order of a few minutes.[RWA15]
Subjects Participants were given time to familiarise themselves with the equipment and testing protocol before data collection began. 3D[RWA16] Ooptical motion (Vicon Motion Systems, Oxford, UK), plantar pressure (Novel GmbH, Munich, Germany) and forceplate (Kistler, Winterthur, Switzerland) data were collected for all conditions and used as inputs to the musculoskeletal model (described below). Optical motion and plantar pressure data were recorded at 100Hz and forceplate data was wer[RWA17]e recorded at 1000Hz. Data were recorded continuously while the subjects participants walked over the level or inclined walkway with a minimum of 5 clean strikes of the forceplate when walking in running shoes (“shod walking”) or in running shoes with the orthotic heel wedges (“wedged walking”). Both the walking condition order (shod or wedged walking) and walking incline (level, uphill or downhill) were randomised. Subjects Participants were given time to become used to the heel wedges when going between shod and wedged walking conditions on each incline and were able to walk freely along the level or inclined walkway without targeting the forceplate.
The musculoskeletal model used here has been described elsewhere previously;37 in summary, it is a unilateral model of the lower limb, scaled to subject participant height and weight by optical markers placed on the pelvis and lower limb (Figure 3), with the Achilles insertion, the first metatarsal head and base, the fifth metatarsal head and base and the tip of the second phalanx digitised as virtual landmarks relative to the marker clusters on the foot and hallux. This The measured optical motion data [RWA18]was were [RWA19]used to calculate joint angles and inter-segmental moments using Euler angle decompositions and Newton-Euler equations at the metatarsophalangeal (MTP), ankle, knee and hip joints respectively and a static optimisation routine used to estimate [RWA20]and muscle forces for the 13 muscles crossing the ankle joint and Achilles tendon force is taken as the sum of the triceps surae muscle forces. The knee and hip were modelled with 3 rotational degrees of freedom and the ankle modelled as a saddle with 2 degrees of freedom and the MTP as a hinge with 1 rotational degree of freedom. Inter-segmental moment data are presented in the local segment coordinate frame in which they were calculated[RWA21]. Centre of pressure (CoP) data is are [RWA22]presented as dimensionless values normalised to foot length, defined as the RMS distance from the calcaneus to the second metatarsal head.
Statistical analyses were performed in Matlab using the Statistical Analysis Toolbox (Version 2010b, The Mathworks Inc.). All data was checked for normality using a Kolmogorov-Smirnov test. As the effect of orthotic heel wedges on lower limb mechanics was the parameter of interest, statistical comparisons between shod and wedged walking for each incline individually was performed using paired t-tests, with the level of significance set at 0.05. P-values under 0.1 are presented as trend changes and values above 0.1 are not presented. Statistical comparisons for all hip, knee, ankle and MTP kinematics and inter-segmental moments were performed at heel-strike (HS), defined by a forceplate force exceeding 40N, weight-acceptance (WA), push-off (PO) and toe-off (TO), with the latter three time-points defined according to changes in knee flexion angle.38 Inter-segmental moments were normalised to body weight (BW) and height (ht).39 Peak muscle forces were compared across all of stance phase, and in the case of the inv/evertor muscles, peak forces were compared during early (<50 %) and late (≥ 50 %) stance.
Results
For the full gait analysis results, the reader is directed to the supplementary material. The results presented here are those directly related to the calculation of the ankle muscle forces. [RWA23] Changes in overall gait dynamics due to heel wedges were only observed during inclined walking as an increase in stance time (Table 1). During inclined walking, delays of 1-2 % stance to the first ground reaction force (GRF) peak when walking with heel wedges was observed (P = .014 and P = .015 for uphill and downhill respectively). During uphill walking an increase in stance time of 20ms was observed (P = .025) along with delays of 1 – 2 % stance to both GRF peaks (P = .014 and P = .028 respectively). Also reductions in the magnitude of the first GRF peak (1 % BW, P = .027) and rate of force development (ROFD) (6% reduction, P = .026) were observed during wedged uphill walking[RWA24]. Compared to shod walking, the most anterior centre of pressure (CoP) position was found to be less anterior during uphill and downhill wedged walking by 3% (P = .002) and 4% (P = .014) foot length respectively.
At all time points considered no changes in peak frontal plane kinematics were observed during level or inclined wedged walking, but changes in sagittal plane ankle kinematics were observed during inclined walking (Figure 4). Flexion-extension angles at the ankle and MTP joints were largely unaffected due to heel wedges during level walking, with only the MTP range of motion (ROM) decreasing significantly (17.7±5.4° vs. 15.4±3.5°, P = .048). During inclined walking, the only c[RWA25]Changes in lower limb kinematicsankle angle were confined to the ankle during early stance, with a more plantar-flexed ankle at HS (-1.4±5.5° vs. -4.5±7.2°, P = .024 and -9.0±4.3° vs. -10.9±5.3°, P = .015 uphill and downhill respectively) and WA (-1.0±5.5° vs. -5.2±9.4°, P = .034, -18.6±4.8° vs. -21.4±5.3°, P = .003 uphill and downhill respectively) during wedged walking. Sagittal plane ankle angles were unaffected by heel wedges during level walking. During downhill walking, there was greater knee flexion at WA (-28.1±5.8° vs. 30.0±5.5°, P = .020[RWA26]). Compared to the shod condition, ankle ROM was observed to decrease during uphill wedged walking (35.1±5.7° vs. 32.1±4.6°, P = .035), but increase during downhill wedged walking (21.9±5.4° vs. 25.9±6.3°, P < .001).
Knee, aAnkle and MTP joint moments were significantly affected by the presence of heel wedges at each incline (Figure 5). Changes at the ankle during level walking were less apparent compared to uphill or downhill walking, with delays to peak ankle dorsi-flexion and plantar-flexion moments by 2% stance and 1% stance respectively only. However, a large increase in knee push-off moment was observed (0.10 vs. -0.00 N m∙BW-1∙ht-1, P = .016) during level wedged walking. [RWA27]
During uphill walking, peak ankle dorsi-flexion moment increased (0.01 vs. 0.06 N m∙BW-1∙ht-1, P = .005), peak plantar flexion moment decreased (-0.85 vs. -0.79 N m∙BW-1∙ht-1, P = .002) and peak inversion moment decreased (0.09 vs. 0.06 N m∙BW-1∙ht-1, P < .001). A decrease in MTP flexion moment was also observed (-0.10 vs. -0.09 N m∙BW-1∙ht-1, P < .001). At the knee, the first adduction peak moment was increased during wedged walking (0.16 vs. 0.20 N m∙BW-1∙ht-1, P < .001), while knee extensor moment at PO decreased (0.16 vs. 0.09 N m∙BW-1∙ht-1, P < .001).[RWA28]
Similar changes in ankle moments were observed during downhill walking, with an increase in dorsiflexion moment (0.09 vs 0.14 N m∙BW-1∙ht-1, P < .001), a decrease in plantar- flexion moment (-0.62 vs. -0.57 N m∙BW-1∙ht-1, P = .002) and a decrease in inversion moment (0.09 vs. 0.07 N m∙BW-1∙ht-1, P = .030). Knee extensor moment was increased at WA (-0.64 vs. -0.71 N m∙BW-1∙ht-1), PO (-0.27 vs. -0.33 N m∙BW-1∙ht-1) and at TO (-0.39 vs. -0.47 N m∙BW-1∙ht-1) during wedged walking (P < .001 for all three time-points).[RWA29]
The most consistent change in muscle force estimates due to heel wedges were in the ankle dorsi-flexors (12–26 % BW increases in peak Tibialis Anterior force) and toe extensors (range of 4–6 % BW increase for Extensor Digitorum/Hallucis Longus forces) muscle forces during the first half of stance across all walking conditions (Figure 6 and Table 2). A second consistent observation was the decrease in peak Tibialis Posterior and toe flexor forces during level and inclined wedged walking, although this was only statistically significant during inclined walking (mean decreases of 12-14 % BW for Tibialis Posterior and 9–11 % BW the toe flexor muscles respectively). Critically, there was no statistically significant reduction in peak Achilles force for any walking incline due to heel wedges (range of 5–14 % BW decrease). The only significant changes in triceps surae loading during uphill walking were decreases in the medial parts of the triceps surae (12 % BW and 6 % BW for medial Soleus and medial Gastrocnemius respectively). During downhill walking, only the medial portion of Soleus showed a significant decrease in peak force (9 % BW). Overall ankle joint reaction force was reduced by 29 % BW during downhill wedged walking.
Discussion
The aim of this study was to quantify the effect of heel wedges on lower limb mechanics, specifically ankle joint angles, moments and muscle forces; and relate these to common injuries such as Achilles tendonitis. The main clinical driver behind assessing the effect of orthotics on ankle loading during inclined walking was to determine how effective heel wedges are at reducing tendon load not only on during level walking, but also on inclined surfaces, where Achilles tendon loads are known to be increased. Given the mixed evidence surrounding the use of heel wedges to manage Achilles tendonitis, a broader characterisation of lower limb mechanics due to orthotic heel wedges would provide some insight into how the body may adapt to walking with such an intervention, allowing for improvements regarding injury management.