- Email: [email protected]
New lower-limb gait biomechanical characteristics in individuals with Achilles tendinopathy: A systematic review update
Gait & Posture 62 (2018) 146–156
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Review
New lower-limb gait biomechanical characteristics in individuals with Achilles tendinopathy: A systematic review update
T
⁎
Ikponmwonsa Ogbonmwan , Bhavesh D. Kumar, Bruce Paton Institute of Sport Exercise & Health (ISEH), University College London, London, UK
A R T I C LE I N FO
A B S T R A C T
Keywords: Gait Achilles Tendinopathy Biomechanics Kinematics Kinetics
Background: Variations in lower-limb biomechanics have recurrently been associated as aetiological factors for Achilles tendinopathy. Objective: To update a previous systematic review examining lower-limb gait biomechanics in Achilles tendinopathy. Design: Systematic Review. Data sources: MEDLINE, EMBASE, CINAHL PLUS, SPORTDiscus and PUBMED databases searched from inception to May 2016. Eligibility criteria for selecting studies: Studies investigating adults with Achilles tendinopathy and lower-limb gait biomechanics including kinematics, kinetics, dynamic plantar-pressures, temporospatial parameters and muscle activity. Results: Fourteen studies were identified, involving 836 participants. Three were prospective studies and 11 were case-control designs. Selection and performance bias were high for all studies except the prospective studies, reporting bias was unclear for all studies. Significant effect size reductions in gait speed (d = −0.80), stride length (d = −0.84) and step length (d = −0.80) were calculated in runners with Achilles tendinopathy. Increased effect sizes for ankle eversion (d = 1.08), time to maximum pronation (d = −1.72), calcaneal inversion (d = −1.82) and ankle and hip joint moments were also established. Significant differences in plantar pressures and timing of ground reaction forces were calculated. Individuals with Achilles tendinopathy demonstrated differences in amplitude and timing of several lower-limb muscles, notably reductions in the onset of activity (d = 2.02) and duration of activation (d = 2.11) in the Gluteus Medius of subjects with Achilles tendinopathy. Conclusion: Eighteen new biomechanical characteristics in individuals with Achilles tendinopathy have been established. This review highlights a topic rich in quantity, but generally weak in quality, consequently results should be interpreted cautiously. High powered prospective studies are required to determine causality.
Key points
• Reductions in gait speed, step length, stride length and onset and • •
⁎
timing of gluteal muscle activity exist in runners with Achilles tendinopathy. Greater ankle and hip joint moments exist in individuals with Achilles tendinopathy. Further investigation is required to quantify the association precisely, enabling clinical applications of these characteristics.
Corresponding author. E-mail address: [email protected] (I. Ogbonmwan).
https://doi.org/10.1016/j.gaitpost.2018.03.010 Received 14 July 2017; Received in revised form 3 February 2018; Accepted 5 March 2018 0966-6362/ Crown Copyright © 2018 Published by Elsevier B.V. All rights reserved.
1. Background 1.1. Achilles tendinopathy Achilles tendinopathy is a term that describes pain, swelling and dysfunction of the Achilles tendon [1,2]. Previously the terminology frequently used by clinicians and researchers has been complicating and in cases inaccurate. ‘Achilles Tendinitis’ has previously widely been used to describe Achilles tendon pain despite no evidence of inflammation noted on biopsies [1]. Maffuli et al. have recommended the terms ‘Tendinitis’ and ‘Tendinosis’ (degeneration of the tendon) be only used following confirmation of the condition using histopathology,
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
surgical exploration or radiographic imaging [1]. ‘Achilles tendinopathy’ is now the preferred term accepted by the general consensus prior to definitive verification [3]. Achilles tendinopathy can be broadly classified into insertional and mid-portion. Insertional tendinopathy defined as at the calcaneustendon junction and mid-portion as 2–6 cm proximal to the Achilles tendon insertion [4]. Achilles tendinopathy is one of the most common lower-limb overuse sports injuries. The majority of studies investigating Achilles tendinopathy epidemiology have been focused on athletic individuals. Kvist et al. reported an incidence of Achilles tendon complaints (97% tendinopathy and 3% rupture) of 14% in 3336 athletes [5]. Annual incidence rates of Achilles tendon complaints have been reported as 7% and 9% respectively in elite runners [6,7]. In a large cohort study of 57,725 Dutch patients registered with a general practitioner, De Jonge et al. reported an annual incidence rate of mid-portion Achilles tendinopathy as 1.85 per 1000 patients registered. Crucially only 35% of the cases in the study was related to sports activity [8]. Despite the significant incidence the exact aetiology remains unknown and is considered multi-factorial. Variations in lower-limb biomechanics including lower-limb kinematics, kinetics and muscle activity have been recurrently associated as aetiological factors for Achilles tendinopathy [9,10]. Anatomical malalignment and poor neuromuscular control are thought to result in uneven loading and stress within the Achilles tendon subsequently resulting in microtrauma [11].
Progressing on the last review on this topic completed in 2011 [18], we have identified a number of biomechanical factors in studies prior to 2011 not analysed in the last review; furthermore, we have identified 5 new studies since 2011, including studies on walking gait and additional prospective studies. Collation and critical appraisal of this new evidence is now required. 2. Objective This systematic review update aims to: 1. Identify, evaluate and summarise the new and existing evidence examining lower-limb gait biomechanical factors during walking and running in adults with Achilles tendinopathy and calculate the effect sizes. 3. Methods This systematic review was designed and conducted according to guidelines outlined by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [19]. This review was registered with PROSPERO (Prospective Register of Systematic Reviews) number CRD42016049677 and the protocol can be accessed here: http://www. crd.york.ac.uk/prospero/display_record.asp?ID=CRD42016049677 3.1. Information sources
1.2. Biomechanics MEDLINE (OVID), EMBASE, CINAHL PLUS, SPORTDiscus and PUBMED electronic databases were searched from inception to May 2016 (week 4).
Lower-limb biomechanical studies in the literature can be broadly divided into lower-limb kinematics, kinetics, electromyography and temporospatial characteristics.
3.2. Search strategy 1.2.1. Kinematics Kinematics involves the recording of optical motion data to identify and calculate positions, angles, velocities, and accelerations of body segments and joints. This is often done from multiple planes including frontal, sagittal and transverse [12,13].
A targeted and reproducible literature search of the above databases was conducted by two authors (IO and BK). No filters were applied to capture new articles still in press. The search was restricted to human studies and articles published in English. Search outputs were exported to Endnote X7 (Thomson Reuters, Carlsbad, California, USA) individually and a master file was created including all search outputs. Duplicates were found and removed using the in-built endnote function “find duplicates”. All details of the search including hits of each database, total overall hits and number of duplicates was recorded. Search algorithms were created using pre-defined search terms and Boolean operators. (Appendix A)
1.2.2. Kinetics Gait kinetics is the analysis of the ground reaction forces and plantar pressures produced during movement. Ground reaction forces are measured using fixed force plates, or now more commonly instrumented treadmills (in-built force plates) [14,15]. 1.2.3. Electromyography (EMG) EMG is a technique used to measure electrical skeletal muscle activity, including duration, amplitude and muscle activation timing. Commonly this involves the use of electrodes attached to the overlying skin surface of the tested muscle. In gait studies, EMG can be used to analyse how different muscles are activated during the gait cycle [16].
3.3. Eligibility criteria Articles were screened for eligibility using the criteria below: 3.3.1. Inclusion criteria 3.3.1.1. Participants.
1.2.4. Temporospatial characteristics Temporospatial characteristics are the fundamental parameters of an individual’s gait recorded during the gait cycle. These parameters include the measurement of time and distance of steps and strides and phases of the gait cycle, from initial contact to toe-off and also the calculation of speed including gait speed and cadence (step frequency). These parameters are often accurately acquired using kinematic or kinetic techniques [17].
• Human adults with confirmed (Clinician/Imaging/Histopathology)
Achilles tendinopathy/Tendinitis/Tenosynovitis/Tendinosis/Tenopathy/ Paratenonitis/Peritendinitis/Achiilodynia. (Due to previous terminology to describe Achilles tendinopathy)
3.3.1.2. Interventions.
• Studies investigating the association between Achilles tendinopathy and lower-limb biomechanics during walking or running.
1.3. Justification
3.3.1.3. Comparisons.
Identifying the risk factors and evaluating the strength of the evidence will aid in the understanding of the relationship between gait biomechanics and Achilles tendinopathy and is crucial in the development of improved preventative measures and treatment strategies.
• Healthy control adults with no previous history of Achilles tendinopathy and no concurrent musculoskeletal injuries.
147
Gait & Posture 62 (2018) 146–156
Identification
I. Ogbonmwan et al.
Records identified through database searches
Medline=473 PubMed=455 Embase=705 CINAHL PLUS=255 SPORTDiscus=338
Eligibility
Screening
Duplicate records removed (n =944)
Records screened (n=1282)
Records excluded (n=1261)
Full-text articles assessed for eligibility (n =21)
Full-text articles excluded, with reasons (n =7)
Inclusion
7 articles did not measure biomechanical variables during running or walking. Studies included in Systematic Review (n = 14)
Fig. 1. Flow Chart of Literature Search.
and were assessed against the eligibility criteria by two authors (IO and BK). Any disagreement over eligibility of particular studies was resolved through discussion with a third author (BP) Articles were then either excluded or included for data extraction. The references of included papers were also searched for articles potentially missed during the electronic search.
3.3.1.4. Outcomes.
• Biomechanical gait characteristics including, lower-limb kinematics,
kinetics, dynamic plantar pressures, temporospatial parameters and muscle activity (EMG).
3.3.2. Exclusion criteria 3.3.2.1. Participants.
3.5. Data extraction
• Subjects with concurrent injuries other than Achilles tendinopathy
Following trials of a pilot, a standardised data collection form was developed to extract the required outcome variables. (Appendix B) Two authors (IO and BK) extracted data independently, discrepancies were identified and resolved through discussion with a third author (BP) when necessary. If data was incomplete an attempt was made to correspond the authors for missing data.
(Due to confounding factors)
3.3.2.2. Interventions.
• Studies
on Achilles Tendon Rupture/Surgery/Diagnosis/Animals (Not relevant)
3.6. Risk of bias 3.3.2.3. Study design.
• • • • •
The quality of each included study was assessed using the Cochrane Risk of Bias Tool [20]. This tool assesses the following five domains:
Letters/Opinion articles/Editorials (Due to low level of evidence) Unpublished studies (Absence of peer review process) Reviews (Original articles only) Abstracts (Inadequate information) Non-English articles (Unable to read)
1. Selection bias (random sequence generation and allocation concealment) 2. Performance bias (blinding to which intervention a participant received) 3. Detection bias (blinding of outcome assessment) 4. Attrition bias (completeness of outcome data, including exclusions from analysis) 5. Reporting bias (Complete outcome reporting)
3.4. Study selection Selection of articles were made using a tiered approach. Articles were initially screened independently by analysis of titles and abstracts by two authors (IO and BK) to identify studies that potentially met the inclusion criteria outlined above. The full papers were then obtained 148
42 runners (> 15 km/week), 21(16 M, 5F) with mid-portion Achilles tendinopathy, 21 (16 M, 5F) healthy controls. Age − 41.8 (9.7)
22 experienced runners (gender N.S.), 14 control subjects, 8 with Achilles tendinopathy (type not specified). Age − 36 (9)
60 (40 M, 20F) experienced runners, 30 (20 M, 10F) with midportion Achilles tendinopathy, 30 healthy controls. Age 41- (7) 22 (20 M, 2F) healthy recreational runners, 11 (10 M, 1F) runners with Achilles tendinopathy (Type N.S.) and high level of pronation confirmed by podiatrist prior to study, and 11 (10 M, 1F) controls matched for age-39.6 (7.7), height and weight (not matched for pronation) 12(1F 11 M) individuals with chronic Achilles tendinopathy (8 unilateral, 4 bilateral AT)(Type N.S.) and high level of pronation confirmed by podiatrist prior to study 12 (1F 11 M) controls, matched for age – 44.3 (8.4), height and weight (not matched for foot type or levels of pronation.) All individuals involved in running-based sports 33 male runners (> 20 km/week), 14 with mid-portion Achilles tendinopathy (Confirmed on US), 19 healthy control subjects. Age 43(8) 142 healthy adult runners (> 20 km/week) (18–55) completed the study. 45 runners developed an overuse injury. 10/45 developed Achilles tendinopathy(Type N.S.) (8 M 2F). 10 uninjured controls were matched for gender, BMI and age- 45(5) 449 male SEAL (US special armed forces) trainees. Age 22.5 (2.5). 149 suffered lower-limb overuse injuries. 23/149 developed Achilles tendinopathy (Type N.S.) No sig. diff. in uninjured controls for age-22.5 (2.5), race, height, and weight
Azevedo et al. [10]
Baur et al. [21]
Baur et al. [22]
149
40 runners 20 (10 M 10F) with Achilles tendinopathy confirmed on US. (Type N.S.) and 20 (10 M 10F) healthy controls, matched for age − 27 (4.6) height and weight and running experience
89 recreational and competitive runners, 31 with mid-portion Achilles tendinopathy (gender N.S) 58 healthy controls were matched for age – 38.4(1.8) height and mass
48 male runners ( > 30 km/week), 27 with mid- portion Achilles tendinopathy (US confirmed), 21 healthy controls matched for age-40 (7) height and weight
Kim et al. [26]
McCrory et al. [28]
Ryan et al. [30]
Kaufman et al. [32]
Hein et al. [31]
Franettovich et al. [25]
Donoghue et al. [24]
Donoghue et al. [23]
Participants
Reference
Table 1 Systematic Review Study Characteristics.
Association between biomechanical variables and Achilles tendinopathy during running (self-selected speed) using kinematic shod (non-standardised footwear) treadmill analysis and kinetic force plates Association between biomechanics and Achilles tendinopathy during overground, barefoot running (self-selected speed)
Cross-sectional, case-control
Cross-sectional, case-control
Cross-sectional, case-control
During 25-week training period. Investigation to identify association between biomechanical factors and risk of developing overuse injuries including Achilles tendinopathy. Subjects had foot pressures measured whilst walking over force plates (self-selected speed), barefoot and shod (standardised military boot) Association between biomechanics and Achilles tendinopathy during overground, barefoot walking (self-selected speed)
Prospective cohort
Prospective cohort
Gluteal muscle activity patterns associated with Achilles tendinopathy during overground running (14.4 km/h), in standardized shoes Association between biomechanical variables and risk of Achilles tendinopathy during overground, barefoot running (12 km/h)
Association between biomechanics and Achilles tendinopathy during treadmill running (self-selected speed), shod (nonstandardised footwear) with orthotics and without orthotics
Cross-sectional, case-control
Cross-sectional, case-control
Muscle activity patterns associated with Achilles tendinopathy during treadmill running (12 km/h) with standardised shoes Association between biomechanics and Achilles tendinopathy during treadmill running (self-selected speed), barefoot and shod (no details on footwear)
Cross-sectional, case-control Cross-sectional, case-control
Biomechanical variables associated with Achilles tendinopathy during barefoot treadmill running (12 km/h)
Biomechanical variables associated with Achilles tendinopathy during overground running (self-selected speed) in standardised shoes
Cross-sectional, case-control
Cross-sectional, case-control
Intervention
Study Design
(continued on next page)
Kinematics: frontal and sagittal plane rearfoot and transverse plane tibia
Kinematics: Sagittal plane- Hip, Knee and Ankle joint angles Temporospatial parameters: gait speed, stride length, stride time, stride frequency, step length, step width and double limb support Kinematics: frontal plane rearfoot. Kinetics: antero-posterior, medio-lateral and vertical ground reaction forces
Plantar pressures: dynamic arch index
Kinematics: Sagittal plane- knee and ankle angles. frontal plane- rearfoot angles
Muscle activity (EMG): Gluteus Medius and Gluteus Maximus
Kinematics: frontal plane- rearfoot and lower leg angles, sagittal plane- ankle and knee joint angles (Functional principal component analysis used to analyse kinematics)
Muscle activity (EMG): tibialis anterior, peroneus longus, lateral gastrocnemius, rectus femoris, biceps femoris and gluteus medius Kinematics: sagittal plane hip, knee and ankle joint angles Kinetics: anterior-posterior and vertical ground reaction forces Temporospatial parameters: speed, stride length, stride time, stride frequency Muscle activity (EMG): tibialis anterior, peroneals, lateral gastrocnemius, medial gastrocnemius, soleus. Kinetics: anterior-posterior and vertical ground reaction force. Plantar pressure distribution: deviation of the centre of pressure Muscle activity (EMG): tibialis anterior, peroneus longus and medial gastrocnemius Kinematics: frontal plane- rearfoot and lower leg angles, sagittal plane- ankle and knee joint angles
Outcome Measures
I. Ogbonmwan et al.
Gait & Posture 62 (2018) 146–156
Gait & Posture 62 (2018) 146–156
Mean and standard deviations of variables were extracted from included articles (or obtained from the authors, if not presented) and used to calculate effect sizes (Cohen’s d) with 95% confidence intervals to enable comparison between study results. Cohen's d = (M2 − M1)/ SDpooled (M = mean, SD = Standard deviation) Effect sizes were considered statistically significant if their 95% confidence intervals did not cross 0.
Muscle Activity (EMG): triceps surae muscles
Kinematics: transverse plane, tibial motion and knee motion
4. Results 4.1. Search results A total of 2226 articles were identified through our database searches. A flow chart of the search process is shown below. (Fig. 1) 4.2. Study characteristics The 14 studies included in the systematic review are described in detail below (Table 1) 11 studies utilised a Cross-sectional, Case-control design [10,21–30] and 3 were Prospective cohorts [31–33]. The studies conducted between 1999 and 2015 included a total of 836 participants. Study size ranged from 16 to 323. The study outcome measures are reported in (Table 2) 4.2.1. Participant characteristics All participants in the studies were active. Participants in 12 out of 14 of the studies were specifically runners [10,21,22,24,25, 31,26,28,30,33,29,27]. The mean age of participants was 33 (11). 72.9% of the participants were male and 27.1% were female. Participants in 7 out of the 14 studies had confirmed mid-portion Achilles tendinopathy [10,22,25,28,30,33,27], tendinopathy type was not specified in the other 7 studies [21,23,24,31,32,26,29].
Cross-sectional, case-control
Association between transverse plane kinematics of the knee and tibia in rearfoot strikers with and without Achilles tendinopathy during overground running (12 km/h) (footwear N.S.) Association between neuromotor control of the triceps surae and Achilles tendinopathy during shod(standardized) overground running (14.4 km/h) Cross-sectional, case-control
4.2.2. Intervention characteristics 6 studies conducted biomechanical testing with participants shod [10,22,23,25,28,27] 5 studies in barefoot [21,31,26,30,33], 2 in both barefoot and shod [24,32] and 1 study did not specify [29]. 9 studies measured participants walking overground in the lab [10,25,31,32,26,30,33,29,27] and the other 5 studies were conducted using treadmills [21–24,28]. 4.3. Study quality The Cochrane risk of bias tool was used to assess quality of the studies individually, (Fig. 2) and overall as a percentage. (Fig. 3) None of the studies utilised random sequence generation. Allocation concealment bias and risk of bias from lack of blinding was high for all included studies except the 3 prospective cohort study designs. [31–33] Attrition bias was low for all studies except the 7 studies which did not specify the type of Achilles tendinopathy. Due to no access to study Table 2 Study Intervention Outcomes.
Wyndow et al. [27]
Williams III et al. [29]
Association between foot pressure distribution and risk of Achilles tendinopathy during barefoot overground running (self-selected speed) Prospective cohort
129 (19 M, 110 F) novice runners aged −39(10) in a 10-week Start-to-Run program, 10 (2 M, 8F) developed diagnosed midportion Achilles tendinopathy, 53 (8 M, 45F) with no injuries used as controls 16 runners ( > 9.7 km/week), 8(6 M, 2F) with diagnosed Achilles tendinopathy(Type N.S) (asymptomatic) and 8(5 M 3F) healthy controls, matched for age − 36(8.2), height, weight and running mileage 34 male runners( > 20 km/week), 15 with mid-portion Achilles tendinopathy (US confirmed), 19 healthy controls. Age- 42(7) Van Ginckel et al. [33]
Intervention Study Design
Plantar pressures: temporal data, peak force, force-time integrals, contact time, medio-lateral force ratios and position and deviation of the centre of force
3.7. Statistical analysis
Participants Reference
Table 1 (continued)
Outcome Measures
I. Ogbonmwan et al.
150
Intervention Outcomes
Number of studies
Kinematics EMG Kinetics + Kinematics Plantar Pressures Kinematics + Temporospatial characteristics EMG + Kinetics + Plantar Pressures EMG + Kinematics + Kinetics + Temporospatial characteristics
5 3 1 2 1 1 1
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Fig. 2. Risk of bias summary: individual studies. + low risk, - high risk,? unclear risk.
Fig. 3. Risk of bias graph: Percentages across all included studies.
(RR) of developing Achilles tendinopathy in patients with pes cavus (RR = 1.25) and pes planus (RR = 1.93) [32]. Van Ginckel et al. reported a number of significant differences between individuals with Achilles tendinopathy and healthy controls [33]. (Table 4) (Fig. 5)
protocols it was unclear for all included studies whether there was selective reporting. 4.4. Results of individual studies 4.4.1. Temporospatial gait characteristics 2 out of the 14 studies examining temporospatial gait characteristics reported contrasting results. Azevedo et al. reported no differences [10], whereas Kim et al [26]. (Table 3) (Fig. 4) reported significant reductions in gait speed, stride length and step length in subjects with Achilles tendinopathy compared to healthy individuals.
4.4.3. EMG activity 5 out of the 14 studies investigated muscle activity and timing. One study reported no significant differences in any of the tested muscles. (Tibialis anterior, peroneals, gastrocnemii, soleus)[21]. Four studies reported significant differences between individuals with Achilles tendinopathy and health controls. (Table 5) (Fig. 6)
4.4.2. Plantar pressures 3 out of the 14 studies investigated dynamic gait plantar pressures. Baur et al. [21] reported no significant differences in plantar pressures between individuals with Achilles tendinopathy and healthy controls [21]. Kaufman et al. reported an increased relative risk ratio Table 3 Temporospatial Gait Characteristics. Author
Variable
Control mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Kim et al. [26]
Gait speed (m/s) Stride Length (m) Step length (m)
1.23 (0.17)
1.09 (0.18) 1.17 (0.12) 0.58 (0.07)
−0.80
−1.41, −0.16a −1.46, −0.20a −1.41, −0.16a
1.28 (0.14) 0.64 (0.08)
−0.84 −0.80
Fig. 4. Temporospatial variables Effect Sizes.
151
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Table 4 Plantar Pressure Variables. Author
Variable
Control mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Van Ginckel et al. [33]
Ratio 2 FFF dytotal (mm) dyFFPOP (mm) dyLFC (mm)
−0.168 (0.11) 247.4 (14.43) 63.45 (9.98) 247.4 (14.45)
−0.273 (0.13) 233.8 (13.27) 55.9 (10.41) 233.8 (13.31)
−0.93 −0.95 −0.75 −0.95
−1.61, −1.64, −1.43, −1.63,
−0.22a −0.25a −0.05a −0.24a
Variables: Ratio 2 FFF: force distribution beneath the forefoot at forefoot flat. dytotal: total posterior-anterior displacement of the centre of force. dyFFPOP: total posterior-anterior displacement of the centre of force during the forefoot push-off phase. dyLFC: total posterior-anterior displacement of the centre of force at last foot contact. a statistically significant.
gait biomechanical factors during walking and running in adults with Achilles tendinopathy. Fourteen studies were identified and effect sizes were calculated. 5.1. Principal findings in context of other studies 5.1.1. Temporospatial gait characteristics Lower-limb injury has previously been reported to result in a reduction in temporospatial gait characteristics [33] This is consistent with our findings, in which we noted significant reductions in gait speed (d = −0.80), stride length (d = −0.84) and step length (d = −0.80) in runners with Achilles tendinopathy [26]. This reduction in temporospatial gait characteristics is likely due to a strategic protective and compensatory mechanism.
5. Discussion
5.1.2. Plantar pressures Van Ginckel et al. described a reduced total forward displacement of the centre of force beneath the foot (dytotal) (d = −0.95), a more laterally directed force distribution beneath the forefoot when flat (FFF) (d = −0.93), a reduced forward displacement of the centre of force during the forefoot push-off phase (dyFFPOP) (d = −0.75) and at last foot contact (dyLFC) (d = −0.95) [33]. The increased laterally directed force distribution is in direct contrast to the commonly mentioned hyperpronation of the foot following an inverted heel strike [28]. Van Ginckel hypothesised that this lateral displacement results in increased unveven stress on the lateral side of the Achilles tendon. Furthermore the reduced anterior displacement of the centre of force results in reduced forefoot mobility leading to increased transfer of tensile forces through the Achilles tendon. Conversely Kaufman et al. and Baur et al. [21], noted no significant differences in plantar pressures between groups [21,32]. Notably Van Ginckel et al. is one of the 3 studies with a prospective study design.
The aim of this systematic review update was to identify, evaluate and summarise the new and existing evidence examining lower-limb
5.1.3. EMG Contrasting results were observed in the investigation of the
Fig. 5. Plantar Pressure variables Effect Sizex.
4.4.4. Kinetics 3 out of the 14 studies examined gait kinetics. No significant differences in magnitude or timing of ground reaction forces were reported by Baur et al. [21] and Azevedo et al. [10] between individuals with Achilles tendinopathy and healthy controls. McCrory et al. reported a number of significant variables [28] (Table 6) (Fig. 7) 4.4.5. Kinematics 8 out of the 14 studies examined gait kinematics. (Table 7) (Fig. 8)
Table 5 EMG Activity. Author
Variable
Control Mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Azevedo et al. [10]
Tibialis anterior (pre-heel strike) (%) Rectus femoris (post-heel strike) (%) Gluteus medius (post-heel strike) (%) Peroneus longus (weight acceptance) (mV) Medial gastrocnemius (weight acceptance) (mV) Soleus-lateral gastrocnemius timing (m/s) Gluteus med. onset of activity (s) Gluteus med. duration of activation (s) Gluteus max. onset of activity (s) Gluteus max. duration of activation (s) Gluteus max. offset of activity (s)
22.90 (5.2) 34.10 (8.2) 25.20 (5.4) 2.48 (0.5) 2.78 (0.5) 0.00 (18.2) 0.27 (0.11) 0.41 (0.11) 0.18 (0.09) 0.32 (0.10) 0.14 (0.02)
17.30 (6.0) 21.60 (9.6) 18.10 (7.9) 2.16 (0.7) 2.46 (0.6) −18 (22.1) 0.10 (0.04) 0.23 (0.05) 0.08 (0.04) 0.20 (0.03) 0.12 (0.01)
−1.00 −1.40 −1.05 −0.54 −0.63 0.90 2.02 2.11 1.33 1.47 1.14
−1.62, −0.34a −2.05, −0.70a −1.67, −0.39a −1.04, −0.01a −1.14, −0.10a 1.59, 0.17a 1.990, 2.048a 2.080, 2.138a 1.300, 1.350a 1.444, 1.498a 1.138, 1.151
Baur et al. [22] Wyndow et al. [27] Franettovich et al. [25]
a
Statistically significant.
152
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Fig. 6. EMG Effect Sizes.
Table 6 Kinetic Variables. Author
Variable
Control mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
McCrory et al. [28]
Time to first vertical peak (% stance) Time to min vertical peak (% stance) Second vertical peak force (BW) Time to second vertical force (% stance) Time to max propulsive force (% stance) Average propulsive force (BW) Propulsive impulse (BW.s) Max braking force (BW) Time to max braking force (% stance) Average braking force (BW) Time to max medial force (% stance) Max lateral force (BW) Time to max lateral force (% stance)
12.9 (0.02) 18.4 (0.02) 2.48 (0.23) 45.6 (0.05) 75 (0.02) 0.181 (0.03) 0.022 (0.01) 0.387 (0.07) 23.8 (0.04) 0.186 (0.03) 19.9 (0.12) 0.093 (0.04) 26.4 (0.15)
13.3 (0.03) 18.96 (0.03) 2.62 (0.33) 44.7 (0.03) 74.4 (0.02) −0.179 (0.05) −0.021 (0.05) 0.428 (0.12) 21 (0.07) 0.206 (0.05) 21.4 (0.09) 0.129 (0.09) 24.4 (0.19)
19.54 22.69 0.52 −19.50 −26.49 −9.38 −1.43 0.46 −56.12 0.52 13.25 0.57 −12.05
16.49, 22.24a 19.16, 25.82a 0.07, 0.96a −22.21, −16.46a −30.14, −22.38a −10.73, −7.86a −1.90, −0.93a 0.02, 0.90a −63.81, −47.45a 0.07, 0.96a 11.16, 15.11a 0.12, 1.01a −13.75, −10.13a
BW = force normalized to subject body weight. a Statistically significant.
subjects with Achilles tendinopathy [10]. Franettovich et al. also noted significant reductions in the onset of activity (d = 2.02) and duration of activation (d = 2.11) in Gluteus Medius muscle activity in subjects with Achilles tendinopathy. Franettovich et al. also reported a reduced duration of activation (d = 1.47), delay in the onset of activity
association between neuromuscular electrical activity and Achilles tendinopathy. Baur et al. [21] reported no significant differences [21]. Azevedo et al. observed significant reductions in muscle activity in Tibialis anterior (pre-heel strike), (d = −1.00) Rectus femoris (d = −1.40) and Gluteus Medius (Post-heel strike) (d = −1.05) in
Fig. 7. Kinetic variables Effect Sizes.
153
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Table 7 Kinematic Variables. Author
Variable
Control Mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Azevedo et al. [10] Donoghue et al. [23]
Knee flexion (heel strike-midstance) (°) Leg abduction max angle (barefoot)(°) Leg abduction max angle (shod) (°) Calcaneal angle (shod) (°) Calcaneal angle (barefoot) (°) Max ankle dorsiflexion (°) Max rearfoot eversion (°) Max knee flexion (°) Peak knee internal rotation (°) Tibial external rotation moment (Nm/kg) Ankle eversion ROM (°) Max ankle dorsiflexion velocity (°/s) Ankle ROM in frontal plane (°) Hip mid−stance moment (Nm/kg) Hip terminal swing moment (Nm/kg) Hip pre−swing moment (Nm/kg) Knee pre−swing moment (Nm/kg) Ankle pre-swing moment (Nm/kg) Ankle terminal joint moment (Nm/kg) Time to maximum pronation (% stance) Calcaneus to vertical touchdown angle (°) Ankle dorsiflexion angle (°)
26.3 (3.9) 5.46 (3.61) 5.92 (3.95) 0.82 (5.73) 0.06 (3.21) 14 (5) −3 (4) 41 (4) 7.3 (4.8) –0.80 (0.34) 11 (3) 330 (59) 41 (7) −0.01 (0.24) −0.80 (0.64) −0.37 (0.21) 0.10 (0.09) −0.01 (0.07) −0.03 (0.02) 40.32 (1.38) −6.97 (0.73) Effect Size (d) 0.86 1.08 1.03 0.89
22.0 (5.5) 1.17 (3.78) 2.64 (4.74) −2.94 (6.18) 0.82 (3.05) 9 (3) −5 (3) 37 (7) 3.1 (3.8) –0.33 (0.35) 13 (3) 300 (39) 45 (7) 0.33 (0.32) 0.21 (0.70) −0.10 (0.25) 0.01 (0.08) 0.08 (0.18) 0.00 (0.01) 37.30 (2.31) −8.94 (1.54)
−0.90 −1.16 −0.75 −0.63 0.24 −1.21 −0.57 −0.70 −0.97 1.36 0.67 −0.62 0.57 1.20 1.51 1.17 −1.06 0.66 1.90 −1.72 −1.82
−1.52, −0.25* −2.01, −0.22* −1.59, 0.14 −1.46, 0.25 −0.61, 1.07 −2.11, −0.21* −1.43,0.35 −1.57, 0.23 −1.95, 0.11 0.21, 2.36* 0.07, 1.24* −1.19, −0.02* −0.02, 1.14 0.51, 1.85* 0.78, 2.18* 0.48, 1.81* −1.70, −0.38* 0.01, 1.28* 1.12, 2.60* −2.21, −1.20* −2.32, −1.30* P-Value 0.033* 0.01* 0.062* 0.029*
Hein et al. [31]
Williams III et al. [29] Ryan et al. [30]
Kim et al. [26]
McCrory et al. [28] Donoghue et al. [24]
Ankle eversion angle (°) Calcaneal angle (°) Knee flexion angle (stance phase) (°) (Mean(SD)/CI data not provided by Donoghue et al. [24]). * Statistically significant.
impact on all other aspects of gait biomechanics, resulting in uneven ground reaction forces and plantar pressures, less optimum joint positions and angles and reductions in temporospatial parameters. Addressing this altered neuromotor control in individuals with Achilles tendinopathy should therefore form a more significant aspect of the rehabalitation process.
(d = 1.33) and earlier offset of muscle activity (d = 1.14) in the Gluteus Maximus of subjects with Achilles tendinopathy [25]. This is the only study to date to demonstrate significant differences in muscle activity of Gluteus Maximus in participants with Achilles tendinopathy. Reduced muscle activity of Gluteus Maximus (primary hip extensor during gait) and subsequent reduced hip joint moment has been reported to increase joint moment further along the kinetic chain [34] and thus could be a possible mechanism for the development of Achilles tendinopathy. Baur et al. [22] reported no significant differences between the two groups in muscle activity of Tibialis anterior, however reported reduced muscle activity in the peroneus longus (d = −0.54) and medial gastrocnemius muscles (d = −0.63) during the early stance phase in subjects with Achilles tendinopathy [22]. Wyndow et al. reported an earlier relative offset timing between Soleus and lateral gastrocnemius (d = 0.90) in Achilles tendinopathy subjects, however no significant differences in muscle onset timing [27]. Reductions in muscle activity and delay in muscle timing is likely to have a profound
5.1.4. Kinetics McCrory et al. is the only study to date to investigate timing of ground reaction forces and reported a number of significant differences between individuals with Achilles tendinopathy and health controls. Significant increases in individuals with Achilles tendinopathy were noted for time to first vertical peak (d = 12.9), time to minimum vertical peak (d = 18.4), time to max medial force (d = 13.25). Time to second vertical force (d = −19.50), time to max propulsive force (d = −26.49), time to max braking force (d = −56.12), time to max lateral force (d = −12.05) and average propulsive force (d = −9.38)
Fig. 8. Kinematic variables Effect Sizes.
154
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
5.2. Limitations
were all significantly reduced in individuals with Achilles tendinopathy.[28] As participants were symptomatic during this study, the delay in reaching peak forces and increase in braking forces may likely represent a protective mechanism to minimise further stress to the Achilles tendon.
Limitations exist in the study quality described earlier, particularly study design in which only 3 out of 14 studies utilised a prospective design increasing the difficulty of differentiating between cause and effect. All reported studies were carried out in active people, despite a large cohort study indicating the majority of Achilles tendinopathy is not related to sports activity [8]. Tendinopathy type was not specified in half of the studies making it difficult to understand and establish pathophysiology. The majority of the sample sizes of the included studies were relatively small, possibly distorting effect size calculations between groups. Another important limitation is that the biomechanical variables examined vary significantly between studies making it difficult to make comparisons and thus draw conclusions.
5.1.5. Kinematics Donoghue et al. (1) reported a reduced leg abduction max angle (angle between the lower leg and the ground on the medial side as viewed from the posterior-indicating level of varus/valgus of lower leg) in barefoot individuals with Achilles tendinopathy (d = −1.16) however no significant difference was noted when shod [24]. This reduction in leg abduction correlates with the significant reduction in onset and duration of activation of the Gluteus medius in individuals with Achilles tendinopathy reported by Franettovich et al. [25].
5.3. What this review adds 5.1.5.1. Foot kinematics. 2 out of the 6 studies investigating foot kinematics reported significant differences between healthy individuals and individuals with Achilles tendinopathy. McCrory et al. reporting a significant increase in time to maximum pronation (d = −1.72) and greater calcaneal inversion (increase in calcaneus to vertical touchdown angle) (d = −1.82).[28] Donoghue et al. (2) also reported greater calcaneal inversion at heel-strike (d = 1.03).[23] These findings are in keeping with Clement et al. widely cited excessive pronation following an inverted heel-strike hypothesis [35].
This is the second review of lower-limb gait biomechanics in individuals with Achilles tendinopathy. This study includes data from 5 new studies since the last review [18]. This update identifies and calculates effect sizes for eighteen new significant gait biomechanical characteristics in individuals with Achilles tendinopathy.
5.4. Clinical implications Critical appraisal of the new biomechanical characteristics identified have highlighted the importance of the potential role of gluteal neuromuscular control rehabilitation as a management strategy in Achilles tendinopathy. Important differences in temporospatial characteristics between healthy individuals and those with Achilles tendinopathy require further investigation to quantify the association precisely enabling useful applications of these characteristics including during screening and diagnosis and also as return to play markers during rehabilitation.
5.1.5.2. Ankle kinematics. Donoghue et al. (2) reported increased ankle dorsiflexion in Achilles tendinopathy (d = 0.86) [23]. Contrastingly Hein et al. reported a lower maximum ankle dorsiflexion angle (d = −1.21) [31] and Ryan et al. reported a reduced maximum ankle dorsiflexion velocity (d = 0.62) [30]. Notably Hein et al. is the only study investigating gait kinematics in Achilles tendinopathy with a prospective study design. Donoghue et al. (2) and Ryan et al. both reported a greater ankle eversion angle (d = 1.08) and (d = 0.67) respectively [23,30] and Kim et al. reported an increase in ankle joint moment during the pre-swing phase (d = 0.66) and the terminal swing phase in individuals with Achilles tendinopathy (d = 1.90) [26]. These findings of increased ankle eversion are also in keeping with the excessive pronation ‘whipping action’ hypothesis cited by Clement et al. drawing the Achilles tendon medially leading to microtrauma [35].
6. Conclusion Eighteen new biomechanical characteristics were established in individuals with Achilles tendinopathy, many which have significant potential clinical implications in the prevention and management of Achilles tendinopathy. Runners with Achilles tendinopathy have been reported to have reductions in temporospatial characteristics including stride length, step length and gait speed. Individuals with Achilles tendinopathy have been shown to have increased ankle eversion, ankle and hip joint moments, time to maximum pronation, calcaneal inversion and a decrease in pre-swing phase knee flexion. Individuals with Achilles tendinopathy were also reported to have reductions in muscle activity of tibialis anterior, rectus femoris, peroneus longus, medial gastrocnemius and the gluteal maximus and medius muscles. A reduction in onset of gluteal maximus and medius muscle activity and an increase in offset of soleus and lateral gastrocnemius muscles were also identified. Additionally, altered plantar pressures and differences in the timing of ground reaction forces were established. Clinically we have established the need to emphasise neuromuscular control during the management and rehabilitation process in individuals with Achilles tendinopathy. Due to weak study designs, underpowered studies and inconsistency in reported variables, interpretation of findings must be done with caution. Future studies should focus on high powered prospective studies utilising consistent variables enabling aetiology to be established and subsequently the implementation of preventive and therapeutic measures.
5.1.5.3. Knee kinematics. Azevedo et al. reported reduced knee flexion between heel strike and midstance (d = −0.90) [10], whereas Donoghue et al. (2) reported increased knee flexion (d = 0.89) throughout the entire stance phase in runners with Achilles tendinopathy [23].This correlates with the increased levels of knee flexion observed in Achilles tendinopathy subjects with excessive pronation [36]. Kim et al. reported a significant decrease in the preswing knee moment phase (d = −1.06) in runners with Achilles tendinopathy [26]. This correlates with the reduced hip extensor activity reported by Franettovich et al. resulting in increased forces distally at the Achilles tendon [25]. 5.1.5.4. Hip kinematics. Data on hip joint kinematics is very limited with only data from 2 studies existing. Kim et al. reported a significant increase in hip joint moment for mid-stance (d = 1.20), terminal stance (d = 1.51) and pre-swing phases (d = 1.17) in runners with Achilles tendinopathy [26]. This is in contrast to EMG data by Franettovich et al. reporting reduced hip extensor activity in individuals with Achilles tendinopathy [25]. Azevedo et al. reported no significant differences between healthy individuals and runners with Achilles tendinopathy [10]. 155
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Funding
[17] J. Perry, J.M. Burnfield, Gait Analysis: Normal and Pathological Function, Second Edition. Gait Analysis: Normal and Pathological Function, 2nd ed., Slack Inc, Thorofare, 2010. [18] S.E. Munteanu, C.J. Barton, Lower limb biomechanics during running in individuals with achilles tendinopathy: a systematic review, J. Foot Ankle Res. 4 (2011) 15. [19] A. Liberati, D.G. Altman, J. Tetzlaff, C. Mulrow, P.C. Gotzsche, J.P. Ioannidis, et al., The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration, J. Clin. Epidemiol. 62 (10) (2009) e1–34. [20] J.P.T. Higgins, D.G. Altman, P.C. Gotzsche, P. Juni, D. Moher, A.D. Oxman, et al., The Cochrane Collaboration's tool for assessing risk of bias in randomised trials, Br. Med. J. 343 (2011) 9. [21] H. Baur, C. Divert, A. Hirschmüller, S. Müller, A. Belli, F. Mayer, Analysis of gait differences in healthy runners and runners with chronic Achilles tendon complaints, Isokinet. Exerc. Sci. 12 (2) (2004) 111–116. [22] H. Baur, S. Muller, A. Hirschmuller, M. Cassel, J. Weber, F. Mayer, Comparison in lower leg neuromuscular activity between runners with unilateral mid-portion Achilles tendinopathy and healthy individuals, J. Electromyogr. Kinesiol. 21 (3) (2011) 499–505. [23] O.A. Donoghue, A.J. Harrison, N. Coffey, K. Hayes, Functional data analysis of running kinematics in chronic Achilles tendon injury, Med. Sci. Sports Exerc. 40 (7) (2008) 1323–1335. [24] O.A. Donoghue, A.J. Harrison, P. Laxton, R.K. Jones, Lower limb kinematics of subjects with chronic achilles tendon injury during running, Res. Sports Med. 16 (1) (2008) 23–38. [25] M.M. Franettovich Smith, C. Honeywill, N. Wyndow, K.M. Crossley, M.W. Creaby, Neuromotor control of gluteal muscles in runners with achilles tendinopathy, Med. Sci. Sports Exerc. 46 (3) (2014) 594–599. [26] S. Kim, J. Yu, Changes of gait parameters and lower limb dynamics in recreational runners with achilles tendinopathy, J. Sports Sci. Med. (2015) 1303–2968. [27] N. Wyndow, S.M. Cowan, T.V. Wrigley, K.M. Crossley, Triceps surae activation is altered in male runners with Achilles tendinopathy, J. Electromyogr. Kinesiol. 23 (1) (2013) 166–172. [28] J.L. McCrory, D.F. Martin, R.B. Lowery, D.W. Cannon, W.W. Curl, H.M. Read Jr.et al., Etiologic factors associated with Achilles tendinitis in runners, Med. Sci. Sports Exerc. 31 (10) (1999) 1374–1381. [29] I.D.S.B. Williams, J.A. Zambardino, V.A. Banning, Transverse-plane mechanics at the knee and tibia in runners with and without a history of achilles tendonopathy, J. Orthop. Sports Phys. Ther. 38 (12) (2008) 761–767. [30] M. Ryan, S. Grau, I. Krauss, C. Maiwald, J. Taunton, T. Horstmann, Kinematic analysis of runners with achilles mid-Portion tendinopathy, Foot Ankle Int. 30 (12) (2009) 1190–1195. [31] T. Hein, P. Janssen, U. Wagner-Fritz, G. Haupt, S. Grau, Prospective analysis of intrinsic and extrinsic risk factors on the development of Achilles tendon pain in runners, Scand. J. Med. Sci. Sports 24 (3) (2014) e201-121p. [32] K.R. Kaufman, S.K. Brodine, R.A. Shaffer, C.W. Johnson, T.R. Cullison, The effect of foot structure and range of motion on musculoskeletal overuse injuries, Am. J. Sports Med. 27 (5) (1999) 585–593. [33] A. Van Ginckel, Y. Thijs, N.G. Hesar, N. Mahieu, D. De Clercq, P. Roosen, et al., Intrinsic gait-related risk factors for Achilles tendinopathy in novice runners: a prospective study, Gait Posture 29 (3) (2009) 387–391. [34] H. Sadeghi, S. Sadeghi, F. Prince, P. Allard, H. Labelle, C.L. Vaughan, Functional roles of ankle and hip sagittal muscle moments in able-bodied gait, Clin. Biomech. 16 (8) (2001) 688–695. [35] D.B. Clement, J.E. Taunton, G.W. Smart, Achilles tendinitis and peritendinitis: etiology and treatment, Am. J. Sports Med. 12 (3) (1984) 179–184. [36] I. McClay, K. Manal, A comparison of three-dimensional lower extremity kinematics during running between excessive pronators and normals, Clin. Biomech. 13 (3) (1998) 195–203.
None obtained. Conflicting interests None to be declared. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.gaitpost.2018.03.010. References [1] N. Maffulli, K.M. Khan, G. Puddu, Overuse tendon conditions: time to change a confusing terminology, Arthroscopy: J. Arthrosc. Relat. Surg. 14 (8) (1998) 840–843. [2] N. Maffulli, D. Kader, Tendinopathy of tendo achillis, J. Bone Joint Surg. [Br] 84 (1) (2002) 1–8. [3] M. Paavola, P. Kannus, T.A.H. Jarvinen, K. Khan, L. Jozsa, M. Jarvinen, Current concepts review achilles tendinopathy, J. Bone Joint Surg. [Am] 84A (11) (2002) 2062–2076. [4] X.F. Courville, M.P. Coe, P.J. Hecht, Current concepts review: noninsertional Achilles tendinopathy, Foot Ankle Int. 30 (11) (2009) 1132–1142. [5] M. Kvist, Achilles tendon injuries in athletes, Sports Med. 18 (3) (1994) 173–201. [6] F.E. Johannsen, A.N. Gam, [Achillodynia is not just a sports injury], Ugeskr. Laeger 172 (48) (2010) 3325–3329. [7] J. Lysholm, J. Wiklander, Injuries in runners, Am. J. Sports Med. 15 (2) (1987) 168–171. [8] S. de Jonge, C. van den Berg, R.J. de Vos, H.J. van der Heide, A. Weir, J.A. Verhaar, et al., Incidence of midportion Achilles tendinopathy in the general population, Br. J. Sports Med. 45 (13) (2011) 1026–1028. [9] A.V. Lorimer, P.A. Hume, Achilles tendon injury risk factors associated with running, Sports Med. 44 (10) (2014) 1459–1472. [10] L.B. Azevedo, M.I. Lambert, C.L. Vaughan, C.M. O’Connor, M.P. Schwellnus, Biomechanical variables associated with Achilles tendinopathy in runners, Br. J. Sports Med. 43 (4) (2009) 288–292. [11] N. Wyndow, S.M. Cowan, T.V. Wrigley, K.M. Crossley, Neuromotor control of the lower limb in Achilles tendinopathy: implications for foot orthotic therapy, Sports Med. 40 (9) (2010) 715–727. [12] H.K. Vincent, D.C. Herman, L. Lear-Barnes, R. Barnes, C. Chen, S. Greenberg, et al., Setting standards for medically-based running analysis, Curr. Sports Med. Rep. 13 (4) (2014) 275–283. [13] E. Ceseracciu, Z. Sawacha, C. Cobelli, Comparison of markerless and marker-based motion capture technologies through simultaneous data collection during gait: proof of concept, PLoS One 9 (3) (2014) e87640. [14] F. Dierick, M. Penta, D. Renaut, C. Detrembleur, A force measuring treadmill in clinical gait analysis, Gait Posture 20 (3) (2004) 299–303. [15] P.O. Riley, G. Paolini, U. Della Croce, K.W. Paylo, D.C. Kerrigan, A kinematic and kinetic comparison of overground and treadmill walking in healthy subjects, Gait Posture 26 (1) (2007) 17–24. [16] J.J. Rechtien, J.B. Gelblum, A.J. Haig, A.J. Gitter, Technology assessment: dynamic electromyography in gait and motion analysis, Muscle Nerve 19 (3) (1996) 396–402.
156
Contents lists available at ScienceDirect
Gait & Posture journal homepage: www.elsevier.com/locate/gaitpost
Review
New lower-limb gait biomechanical characteristics in individuals with Achilles tendinopathy: A systematic review update
T
⁎
Ikponmwonsa Ogbonmwan , Bhavesh D. Kumar, Bruce Paton Institute of Sport Exercise & Health (ISEH), University College London, London, UK
A R T I C LE I N FO
A B S T R A C T
Keywords: Gait Achilles Tendinopathy Biomechanics Kinematics Kinetics
Background: Variations in lower-limb biomechanics have recurrently been associated as aetiological factors for Achilles tendinopathy. Objective: To update a previous systematic review examining lower-limb gait biomechanics in Achilles tendinopathy. Design: Systematic Review. Data sources: MEDLINE, EMBASE, CINAHL PLUS, SPORTDiscus and PUBMED databases searched from inception to May 2016. Eligibility criteria for selecting studies: Studies investigating adults with Achilles tendinopathy and lower-limb gait biomechanics including kinematics, kinetics, dynamic plantar-pressures, temporospatial parameters and muscle activity. Results: Fourteen studies were identified, involving 836 participants. Three were prospective studies and 11 were case-control designs. Selection and performance bias were high for all studies except the prospective studies, reporting bias was unclear for all studies. Significant effect size reductions in gait speed (d = −0.80), stride length (d = −0.84) and step length (d = −0.80) were calculated in runners with Achilles tendinopathy. Increased effect sizes for ankle eversion (d = 1.08), time to maximum pronation (d = −1.72), calcaneal inversion (d = −1.82) and ankle and hip joint moments were also established. Significant differences in plantar pressures and timing of ground reaction forces were calculated. Individuals with Achilles tendinopathy demonstrated differences in amplitude and timing of several lower-limb muscles, notably reductions in the onset of activity (d = 2.02) and duration of activation (d = 2.11) in the Gluteus Medius of subjects with Achilles tendinopathy. Conclusion: Eighteen new biomechanical characteristics in individuals with Achilles tendinopathy have been established. This review highlights a topic rich in quantity, but generally weak in quality, consequently results should be interpreted cautiously. High powered prospective studies are required to determine causality.
Key points
• Reductions in gait speed, step length, stride length and onset and • •
⁎
timing of gluteal muscle activity exist in runners with Achilles tendinopathy. Greater ankle and hip joint moments exist in individuals with Achilles tendinopathy. Further investigation is required to quantify the association precisely, enabling clinical applications of these characteristics.
Corresponding author. E-mail address: [email protected] (I. Ogbonmwan).
https://doi.org/10.1016/j.gaitpost.2018.03.010 Received 14 July 2017; Received in revised form 3 February 2018; Accepted 5 March 2018 0966-6362/ Crown Copyright © 2018 Published by Elsevier B.V. All rights reserved.
1. Background 1.1. Achilles tendinopathy Achilles tendinopathy is a term that describes pain, swelling and dysfunction of the Achilles tendon [1,2]. Previously the terminology frequently used by clinicians and researchers has been complicating and in cases inaccurate. ‘Achilles Tendinitis’ has previously widely been used to describe Achilles tendon pain despite no evidence of inflammation noted on biopsies [1]. Maffuli et al. have recommended the terms ‘Tendinitis’ and ‘Tendinosis’ (degeneration of the tendon) be only used following confirmation of the condition using histopathology,
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
surgical exploration or radiographic imaging [1]. ‘Achilles tendinopathy’ is now the preferred term accepted by the general consensus prior to definitive verification [3]. Achilles tendinopathy can be broadly classified into insertional and mid-portion. Insertional tendinopathy defined as at the calcaneustendon junction and mid-portion as 2–6 cm proximal to the Achilles tendon insertion [4]. Achilles tendinopathy is one of the most common lower-limb overuse sports injuries. The majority of studies investigating Achilles tendinopathy epidemiology have been focused on athletic individuals. Kvist et al. reported an incidence of Achilles tendon complaints (97% tendinopathy and 3% rupture) of 14% in 3336 athletes [5]. Annual incidence rates of Achilles tendon complaints have been reported as 7% and 9% respectively in elite runners [6,7]. In a large cohort study of 57,725 Dutch patients registered with a general practitioner, De Jonge et al. reported an annual incidence rate of mid-portion Achilles tendinopathy as 1.85 per 1000 patients registered. Crucially only 35% of the cases in the study was related to sports activity [8]. Despite the significant incidence the exact aetiology remains unknown and is considered multi-factorial. Variations in lower-limb biomechanics including lower-limb kinematics, kinetics and muscle activity have been recurrently associated as aetiological factors for Achilles tendinopathy [9,10]. Anatomical malalignment and poor neuromuscular control are thought to result in uneven loading and stress within the Achilles tendon subsequently resulting in microtrauma [11].
Progressing on the last review on this topic completed in 2011 [18], we have identified a number of biomechanical factors in studies prior to 2011 not analysed in the last review; furthermore, we have identified 5 new studies since 2011, including studies on walking gait and additional prospective studies. Collation and critical appraisal of this new evidence is now required. 2. Objective This systematic review update aims to: 1. Identify, evaluate and summarise the new and existing evidence examining lower-limb gait biomechanical factors during walking and running in adults with Achilles tendinopathy and calculate the effect sizes. 3. Methods This systematic review was designed and conducted according to guidelines outlined by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [19]. This review was registered with PROSPERO (Prospective Register of Systematic Reviews) number CRD42016049677 and the protocol can be accessed here: http://www. crd.york.ac.uk/prospero/display_record.asp?ID=CRD42016049677 3.1. Information sources
1.2. Biomechanics MEDLINE (OVID), EMBASE, CINAHL PLUS, SPORTDiscus and PUBMED electronic databases were searched from inception to May 2016 (week 4).
Lower-limb biomechanical studies in the literature can be broadly divided into lower-limb kinematics, kinetics, electromyography and temporospatial characteristics.
3.2. Search strategy 1.2.1. Kinematics Kinematics involves the recording of optical motion data to identify and calculate positions, angles, velocities, and accelerations of body segments and joints. This is often done from multiple planes including frontal, sagittal and transverse [12,13].
A targeted and reproducible literature search of the above databases was conducted by two authors (IO and BK). No filters were applied to capture new articles still in press. The search was restricted to human studies and articles published in English. Search outputs were exported to Endnote X7 (Thomson Reuters, Carlsbad, California, USA) individually and a master file was created including all search outputs. Duplicates were found and removed using the in-built endnote function “find duplicates”. All details of the search including hits of each database, total overall hits and number of duplicates was recorded. Search algorithms were created using pre-defined search terms and Boolean operators. (Appendix A)
1.2.2. Kinetics Gait kinetics is the analysis of the ground reaction forces and plantar pressures produced during movement. Ground reaction forces are measured using fixed force plates, or now more commonly instrumented treadmills (in-built force plates) [14,15]. 1.2.3. Electromyography (EMG) EMG is a technique used to measure electrical skeletal muscle activity, including duration, amplitude and muscle activation timing. Commonly this involves the use of electrodes attached to the overlying skin surface of the tested muscle. In gait studies, EMG can be used to analyse how different muscles are activated during the gait cycle [16].
3.3. Eligibility criteria Articles were screened for eligibility using the criteria below: 3.3.1. Inclusion criteria 3.3.1.1. Participants.
1.2.4. Temporospatial characteristics Temporospatial characteristics are the fundamental parameters of an individual’s gait recorded during the gait cycle. These parameters include the measurement of time and distance of steps and strides and phases of the gait cycle, from initial contact to toe-off and also the calculation of speed including gait speed and cadence (step frequency). These parameters are often accurately acquired using kinematic or kinetic techniques [17].
• Human adults with confirmed (Clinician/Imaging/Histopathology)
Achilles tendinopathy/Tendinitis/Tenosynovitis/Tendinosis/Tenopathy/ Paratenonitis/Peritendinitis/Achiilodynia. (Due to previous terminology to describe Achilles tendinopathy)
3.3.1.2. Interventions.
• Studies investigating the association between Achilles tendinopathy and lower-limb biomechanics during walking or running.
1.3. Justification
3.3.1.3. Comparisons.
Identifying the risk factors and evaluating the strength of the evidence will aid in the understanding of the relationship between gait biomechanics and Achilles tendinopathy and is crucial in the development of improved preventative measures and treatment strategies.
• Healthy control adults with no previous history of Achilles tendinopathy and no concurrent musculoskeletal injuries.
147
Gait & Posture 62 (2018) 146–156
Identification
I. Ogbonmwan et al.
Records identified through database searches
Medline=473 PubMed=455 Embase=705 CINAHL PLUS=255 SPORTDiscus=338
Eligibility
Screening
Duplicate records removed (n =944)
Records screened (n=1282)
Records excluded (n=1261)
Full-text articles assessed for eligibility (n =21)
Full-text articles excluded, with reasons (n =7)
Inclusion
7 articles did not measure biomechanical variables during running or walking. Studies included in Systematic Review (n = 14)
Fig. 1. Flow Chart of Literature Search.
and were assessed against the eligibility criteria by two authors (IO and BK). Any disagreement over eligibility of particular studies was resolved through discussion with a third author (BP) Articles were then either excluded or included for data extraction. The references of included papers were also searched for articles potentially missed during the electronic search.
3.3.1.4. Outcomes.
• Biomechanical gait characteristics including, lower-limb kinematics,
kinetics, dynamic plantar pressures, temporospatial parameters and muscle activity (EMG).
3.3.2. Exclusion criteria 3.3.2.1. Participants.
3.5. Data extraction
• Subjects with concurrent injuries other than Achilles tendinopathy
Following trials of a pilot, a standardised data collection form was developed to extract the required outcome variables. (Appendix B) Two authors (IO and BK) extracted data independently, discrepancies were identified and resolved through discussion with a third author (BP) when necessary. If data was incomplete an attempt was made to correspond the authors for missing data.
(Due to confounding factors)
3.3.2.2. Interventions.
• Studies
on Achilles Tendon Rupture/Surgery/Diagnosis/Animals (Not relevant)
3.6. Risk of bias 3.3.2.3. Study design.
• • • • •
The quality of each included study was assessed using the Cochrane Risk of Bias Tool [20]. This tool assesses the following five domains:
Letters/Opinion articles/Editorials (Due to low level of evidence) Unpublished studies (Absence of peer review process) Reviews (Original articles only) Abstracts (Inadequate information) Non-English articles (Unable to read)
1. Selection bias (random sequence generation and allocation concealment) 2. Performance bias (blinding to which intervention a participant received) 3. Detection bias (blinding of outcome assessment) 4. Attrition bias (completeness of outcome data, including exclusions from analysis) 5. Reporting bias (Complete outcome reporting)
3.4. Study selection Selection of articles were made using a tiered approach. Articles were initially screened independently by analysis of titles and abstracts by two authors (IO and BK) to identify studies that potentially met the inclusion criteria outlined above. The full papers were then obtained 148
42 runners (> 15 km/week), 21(16 M, 5F) with mid-portion Achilles tendinopathy, 21 (16 M, 5F) healthy controls. Age − 41.8 (9.7)
22 experienced runners (gender N.S.), 14 control subjects, 8 with Achilles tendinopathy (type not specified). Age − 36 (9)
60 (40 M, 20F) experienced runners, 30 (20 M, 10F) with midportion Achilles tendinopathy, 30 healthy controls. Age 41- (7) 22 (20 M, 2F) healthy recreational runners, 11 (10 M, 1F) runners with Achilles tendinopathy (Type N.S.) and high level of pronation confirmed by podiatrist prior to study, and 11 (10 M, 1F) controls matched for age-39.6 (7.7), height and weight (not matched for pronation) 12(1F 11 M) individuals with chronic Achilles tendinopathy (8 unilateral, 4 bilateral AT)(Type N.S.) and high level of pronation confirmed by podiatrist prior to study 12 (1F 11 M) controls, matched for age – 44.3 (8.4), height and weight (not matched for foot type or levels of pronation.) All individuals involved in running-based sports 33 male runners (> 20 km/week), 14 with mid-portion Achilles tendinopathy (Confirmed on US), 19 healthy control subjects. Age 43(8) 142 healthy adult runners (> 20 km/week) (18–55) completed the study. 45 runners developed an overuse injury. 10/45 developed Achilles tendinopathy(Type N.S.) (8 M 2F). 10 uninjured controls were matched for gender, BMI and age- 45(5) 449 male SEAL (US special armed forces) trainees. Age 22.5 (2.5). 149 suffered lower-limb overuse injuries. 23/149 developed Achilles tendinopathy (Type N.S.) No sig. diff. in uninjured controls for age-22.5 (2.5), race, height, and weight
Azevedo et al. [10]
Baur et al. [21]
Baur et al. [22]
149
40 runners 20 (10 M 10F) with Achilles tendinopathy confirmed on US. (Type N.S.) and 20 (10 M 10F) healthy controls, matched for age − 27 (4.6) height and weight and running experience
89 recreational and competitive runners, 31 with mid-portion Achilles tendinopathy (gender N.S) 58 healthy controls were matched for age – 38.4(1.8) height and mass
48 male runners ( > 30 km/week), 27 with mid- portion Achilles tendinopathy (US confirmed), 21 healthy controls matched for age-40 (7) height and weight
Kim et al. [26]
McCrory et al. [28]
Ryan et al. [30]
Kaufman et al. [32]
Hein et al. [31]
Franettovich et al. [25]
Donoghue et al. [24]
Donoghue et al. [23]
Participants
Reference
Table 1 Systematic Review Study Characteristics.
Association between biomechanical variables and Achilles tendinopathy during running (self-selected speed) using kinematic shod (non-standardised footwear) treadmill analysis and kinetic force plates Association between biomechanics and Achilles tendinopathy during overground, barefoot running (self-selected speed)
Cross-sectional, case-control
Cross-sectional, case-control
Cross-sectional, case-control
During 25-week training period. Investigation to identify association between biomechanical factors and risk of developing overuse injuries including Achilles tendinopathy. Subjects had foot pressures measured whilst walking over force plates (self-selected speed), barefoot and shod (standardised military boot) Association between biomechanics and Achilles tendinopathy during overground, barefoot walking (self-selected speed)
Prospective cohort
Prospective cohort
Gluteal muscle activity patterns associated with Achilles tendinopathy during overground running (14.4 km/h), in standardized shoes Association between biomechanical variables and risk of Achilles tendinopathy during overground, barefoot running (12 km/h)
Association between biomechanics and Achilles tendinopathy during treadmill running (self-selected speed), shod (nonstandardised footwear) with orthotics and without orthotics
Cross-sectional, case-control
Cross-sectional, case-control
Muscle activity patterns associated with Achilles tendinopathy during treadmill running (12 km/h) with standardised shoes Association between biomechanics and Achilles tendinopathy during treadmill running (self-selected speed), barefoot and shod (no details on footwear)
Cross-sectional, case-control Cross-sectional, case-control
Biomechanical variables associated with Achilles tendinopathy during barefoot treadmill running (12 km/h)
Biomechanical variables associated with Achilles tendinopathy during overground running (self-selected speed) in standardised shoes
Cross-sectional, case-control
Cross-sectional, case-control
Intervention
Study Design
(continued on next page)
Kinematics: frontal and sagittal plane rearfoot and transverse plane tibia
Kinematics: Sagittal plane- Hip, Knee and Ankle joint angles Temporospatial parameters: gait speed, stride length, stride time, stride frequency, step length, step width and double limb support Kinematics: frontal plane rearfoot. Kinetics: antero-posterior, medio-lateral and vertical ground reaction forces
Plantar pressures: dynamic arch index
Kinematics: Sagittal plane- knee and ankle angles. frontal plane- rearfoot angles
Muscle activity (EMG): Gluteus Medius and Gluteus Maximus
Kinematics: frontal plane- rearfoot and lower leg angles, sagittal plane- ankle and knee joint angles (Functional principal component analysis used to analyse kinematics)
Muscle activity (EMG): tibialis anterior, peroneus longus, lateral gastrocnemius, rectus femoris, biceps femoris and gluteus medius Kinematics: sagittal plane hip, knee and ankle joint angles Kinetics: anterior-posterior and vertical ground reaction forces Temporospatial parameters: speed, stride length, stride time, stride frequency Muscle activity (EMG): tibialis anterior, peroneals, lateral gastrocnemius, medial gastrocnemius, soleus. Kinetics: anterior-posterior and vertical ground reaction force. Plantar pressure distribution: deviation of the centre of pressure Muscle activity (EMG): tibialis anterior, peroneus longus and medial gastrocnemius Kinematics: frontal plane- rearfoot and lower leg angles, sagittal plane- ankle and knee joint angles
Outcome Measures
I. Ogbonmwan et al.
Gait & Posture 62 (2018) 146–156
Gait & Posture 62 (2018) 146–156
Mean and standard deviations of variables were extracted from included articles (or obtained from the authors, if not presented) and used to calculate effect sizes (Cohen’s d) with 95% confidence intervals to enable comparison between study results. Cohen's d = (M2 − M1)/ SDpooled (M = mean, SD = Standard deviation) Effect sizes were considered statistically significant if their 95% confidence intervals did not cross 0.
Muscle Activity (EMG): triceps surae muscles
Kinematics: transverse plane, tibial motion and knee motion
4. Results 4.1. Search results A total of 2226 articles were identified through our database searches. A flow chart of the search process is shown below. (Fig. 1) 4.2. Study characteristics The 14 studies included in the systematic review are described in detail below (Table 1) 11 studies utilised a Cross-sectional, Case-control design [10,21–30] and 3 were Prospective cohorts [31–33]. The studies conducted between 1999 and 2015 included a total of 836 participants. Study size ranged from 16 to 323. The study outcome measures are reported in (Table 2) 4.2.1. Participant characteristics All participants in the studies were active. Participants in 12 out of 14 of the studies were specifically runners [10,21,22,24,25, 31,26,28,30,33,29,27]. The mean age of participants was 33 (11). 72.9% of the participants were male and 27.1% were female. Participants in 7 out of the 14 studies had confirmed mid-portion Achilles tendinopathy [10,22,25,28,30,33,27], tendinopathy type was not specified in the other 7 studies [21,23,24,31,32,26,29].
Cross-sectional, case-control
Association between transverse plane kinematics of the knee and tibia in rearfoot strikers with and without Achilles tendinopathy during overground running (12 km/h) (footwear N.S.) Association between neuromotor control of the triceps surae and Achilles tendinopathy during shod(standardized) overground running (14.4 km/h) Cross-sectional, case-control
4.2.2. Intervention characteristics 6 studies conducted biomechanical testing with participants shod [10,22,23,25,28,27] 5 studies in barefoot [21,31,26,30,33], 2 in both barefoot and shod [24,32] and 1 study did not specify [29]. 9 studies measured participants walking overground in the lab [10,25,31,32,26,30,33,29,27] and the other 5 studies were conducted using treadmills [21–24,28]. 4.3. Study quality The Cochrane risk of bias tool was used to assess quality of the studies individually, (Fig. 2) and overall as a percentage. (Fig. 3) None of the studies utilised random sequence generation. Allocation concealment bias and risk of bias from lack of blinding was high for all included studies except the 3 prospective cohort study designs. [31–33] Attrition bias was low for all studies except the 7 studies which did not specify the type of Achilles tendinopathy. Due to no access to study Table 2 Study Intervention Outcomes.
Wyndow et al. [27]
Williams III et al. [29]
Association between foot pressure distribution and risk of Achilles tendinopathy during barefoot overground running (self-selected speed) Prospective cohort
129 (19 M, 110 F) novice runners aged −39(10) in a 10-week Start-to-Run program, 10 (2 M, 8F) developed diagnosed midportion Achilles tendinopathy, 53 (8 M, 45F) with no injuries used as controls 16 runners ( > 9.7 km/week), 8(6 M, 2F) with diagnosed Achilles tendinopathy(Type N.S) (asymptomatic) and 8(5 M 3F) healthy controls, matched for age − 36(8.2), height, weight and running mileage 34 male runners( > 20 km/week), 15 with mid-portion Achilles tendinopathy (US confirmed), 19 healthy controls. Age- 42(7) Van Ginckel et al. [33]
Intervention Study Design
Plantar pressures: temporal data, peak force, force-time integrals, contact time, medio-lateral force ratios and position and deviation of the centre of force
3.7. Statistical analysis
Participants Reference
Table 1 (continued)
Outcome Measures
I. Ogbonmwan et al.
150
Intervention Outcomes
Number of studies
Kinematics EMG Kinetics + Kinematics Plantar Pressures Kinematics + Temporospatial characteristics EMG + Kinetics + Plantar Pressures EMG + Kinematics + Kinetics + Temporospatial characteristics
5 3 1 2 1 1 1
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Fig. 2. Risk of bias summary: individual studies. + low risk, - high risk,? unclear risk.
Fig. 3. Risk of bias graph: Percentages across all included studies.
(RR) of developing Achilles tendinopathy in patients with pes cavus (RR = 1.25) and pes planus (RR = 1.93) [32]. Van Ginckel et al. reported a number of significant differences between individuals with Achilles tendinopathy and healthy controls [33]. (Table 4) (Fig. 5)
protocols it was unclear for all included studies whether there was selective reporting. 4.4. Results of individual studies 4.4.1. Temporospatial gait characteristics 2 out of the 14 studies examining temporospatial gait characteristics reported contrasting results. Azevedo et al. reported no differences [10], whereas Kim et al [26]. (Table 3) (Fig. 4) reported significant reductions in gait speed, stride length and step length in subjects with Achilles tendinopathy compared to healthy individuals.
4.4.3. EMG activity 5 out of the 14 studies investigated muscle activity and timing. One study reported no significant differences in any of the tested muscles. (Tibialis anterior, peroneals, gastrocnemii, soleus)[21]. Four studies reported significant differences between individuals with Achilles tendinopathy and health controls. (Table 5) (Fig. 6)
4.4.2. Plantar pressures 3 out of the 14 studies investigated dynamic gait plantar pressures. Baur et al. [21] reported no significant differences in plantar pressures between individuals with Achilles tendinopathy and healthy controls [21]. Kaufman et al. reported an increased relative risk ratio Table 3 Temporospatial Gait Characteristics. Author
Variable
Control mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Kim et al. [26]
Gait speed (m/s) Stride Length (m) Step length (m)
1.23 (0.17)
1.09 (0.18) 1.17 (0.12) 0.58 (0.07)
−0.80
−1.41, −0.16a −1.46, −0.20a −1.41, −0.16a
1.28 (0.14) 0.64 (0.08)
−0.84 −0.80
Fig. 4. Temporospatial variables Effect Sizes.
151
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Table 4 Plantar Pressure Variables. Author
Variable
Control mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Van Ginckel et al. [33]
Ratio 2 FFF dytotal (mm) dyFFPOP (mm) dyLFC (mm)
−0.168 (0.11) 247.4 (14.43) 63.45 (9.98) 247.4 (14.45)
−0.273 (0.13) 233.8 (13.27) 55.9 (10.41) 233.8 (13.31)
−0.93 −0.95 −0.75 −0.95
−1.61, −1.64, −1.43, −1.63,
−0.22a −0.25a −0.05a −0.24a
Variables: Ratio 2 FFF: force distribution beneath the forefoot at forefoot flat. dytotal: total posterior-anterior displacement of the centre of force. dyFFPOP: total posterior-anterior displacement of the centre of force during the forefoot push-off phase. dyLFC: total posterior-anterior displacement of the centre of force at last foot contact. a statistically significant.
gait biomechanical factors during walking and running in adults with Achilles tendinopathy. Fourteen studies were identified and effect sizes were calculated. 5.1. Principal findings in context of other studies 5.1.1. Temporospatial gait characteristics Lower-limb injury has previously been reported to result in a reduction in temporospatial gait characteristics [33] This is consistent with our findings, in which we noted significant reductions in gait speed (d = −0.80), stride length (d = −0.84) and step length (d = −0.80) in runners with Achilles tendinopathy [26]. This reduction in temporospatial gait characteristics is likely due to a strategic protective and compensatory mechanism.
5. Discussion
5.1.2. Plantar pressures Van Ginckel et al. described a reduced total forward displacement of the centre of force beneath the foot (dytotal) (d = −0.95), a more laterally directed force distribution beneath the forefoot when flat (FFF) (d = −0.93), a reduced forward displacement of the centre of force during the forefoot push-off phase (dyFFPOP) (d = −0.75) and at last foot contact (dyLFC) (d = −0.95) [33]. The increased laterally directed force distribution is in direct contrast to the commonly mentioned hyperpronation of the foot following an inverted heel strike [28]. Van Ginckel hypothesised that this lateral displacement results in increased unveven stress on the lateral side of the Achilles tendon. Furthermore the reduced anterior displacement of the centre of force results in reduced forefoot mobility leading to increased transfer of tensile forces through the Achilles tendon. Conversely Kaufman et al. and Baur et al. [21], noted no significant differences in plantar pressures between groups [21,32]. Notably Van Ginckel et al. is one of the 3 studies with a prospective study design.
The aim of this systematic review update was to identify, evaluate and summarise the new and existing evidence examining lower-limb
5.1.3. EMG Contrasting results were observed in the investigation of the
Fig. 5. Plantar Pressure variables Effect Sizex.
4.4.4. Kinetics 3 out of the 14 studies examined gait kinetics. No significant differences in magnitude or timing of ground reaction forces were reported by Baur et al. [21] and Azevedo et al. [10] between individuals with Achilles tendinopathy and healthy controls. McCrory et al. reported a number of significant variables [28] (Table 6) (Fig. 7) 4.4.5. Kinematics 8 out of the 14 studies examined gait kinematics. (Table 7) (Fig. 8)
Table 5 EMG Activity. Author
Variable
Control Mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Azevedo et al. [10]
Tibialis anterior (pre-heel strike) (%) Rectus femoris (post-heel strike) (%) Gluteus medius (post-heel strike) (%) Peroneus longus (weight acceptance) (mV) Medial gastrocnemius (weight acceptance) (mV) Soleus-lateral gastrocnemius timing (m/s) Gluteus med. onset of activity (s) Gluteus med. duration of activation (s) Gluteus max. onset of activity (s) Gluteus max. duration of activation (s) Gluteus max. offset of activity (s)
22.90 (5.2) 34.10 (8.2) 25.20 (5.4) 2.48 (0.5) 2.78 (0.5) 0.00 (18.2) 0.27 (0.11) 0.41 (0.11) 0.18 (0.09) 0.32 (0.10) 0.14 (0.02)
17.30 (6.0) 21.60 (9.6) 18.10 (7.9) 2.16 (0.7) 2.46 (0.6) −18 (22.1) 0.10 (0.04) 0.23 (0.05) 0.08 (0.04) 0.20 (0.03) 0.12 (0.01)
−1.00 −1.40 −1.05 −0.54 −0.63 0.90 2.02 2.11 1.33 1.47 1.14
−1.62, −0.34a −2.05, −0.70a −1.67, −0.39a −1.04, −0.01a −1.14, −0.10a 1.59, 0.17a 1.990, 2.048a 2.080, 2.138a 1.300, 1.350a 1.444, 1.498a 1.138, 1.151
Baur et al. [22] Wyndow et al. [27] Franettovich et al. [25]
a
Statistically significant.
152
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Fig. 6. EMG Effect Sizes.
Table 6 Kinetic Variables. Author
Variable
Control mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
McCrory et al. [28]
Time to first vertical peak (% stance) Time to min vertical peak (% stance) Second vertical peak force (BW) Time to second vertical force (% stance) Time to max propulsive force (% stance) Average propulsive force (BW) Propulsive impulse (BW.s) Max braking force (BW) Time to max braking force (% stance) Average braking force (BW) Time to max medial force (% stance) Max lateral force (BW) Time to max lateral force (% stance)
12.9 (0.02) 18.4 (0.02) 2.48 (0.23) 45.6 (0.05) 75 (0.02) 0.181 (0.03) 0.022 (0.01) 0.387 (0.07) 23.8 (0.04) 0.186 (0.03) 19.9 (0.12) 0.093 (0.04) 26.4 (0.15)
13.3 (0.03) 18.96 (0.03) 2.62 (0.33) 44.7 (0.03) 74.4 (0.02) −0.179 (0.05) −0.021 (0.05) 0.428 (0.12) 21 (0.07) 0.206 (0.05) 21.4 (0.09) 0.129 (0.09) 24.4 (0.19)
19.54 22.69 0.52 −19.50 −26.49 −9.38 −1.43 0.46 −56.12 0.52 13.25 0.57 −12.05
16.49, 22.24a 19.16, 25.82a 0.07, 0.96a −22.21, −16.46a −30.14, −22.38a −10.73, −7.86a −1.90, −0.93a 0.02, 0.90a −63.81, −47.45a 0.07, 0.96a 11.16, 15.11a 0.12, 1.01a −13.75, −10.13a
BW = force normalized to subject body weight. a Statistically significant.
subjects with Achilles tendinopathy [10]. Franettovich et al. also noted significant reductions in the onset of activity (d = 2.02) and duration of activation (d = 2.11) in Gluteus Medius muscle activity in subjects with Achilles tendinopathy. Franettovich et al. also reported a reduced duration of activation (d = 1.47), delay in the onset of activity
association between neuromuscular electrical activity and Achilles tendinopathy. Baur et al. [21] reported no significant differences [21]. Azevedo et al. observed significant reductions in muscle activity in Tibialis anterior (pre-heel strike), (d = −1.00) Rectus femoris (d = −1.40) and Gluteus Medius (Post-heel strike) (d = −1.05) in
Fig. 7. Kinetic variables Effect Sizes.
153
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Table 7 Kinematic Variables. Author
Variable
Control Mean (SD)
Case Mean (SD)
Calculated Effect Size (d)
95% CI
Azevedo et al. [10] Donoghue et al. [23]
Knee flexion (heel strike-midstance) (°) Leg abduction max angle (barefoot)(°) Leg abduction max angle (shod) (°) Calcaneal angle (shod) (°) Calcaneal angle (barefoot) (°) Max ankle dorsiflexion (°) Max rearfoot eversion (°) Max knee flexion (°) Peak knee internal rotation (°) Tibial external rotation moment (Nm/kg) Ankle eversion ROM (°) Max ankle dorsiflexion velocity (°/s) Ankle ROM in frontal plane (°) Hip mid−stance moment (Nm/kg) Hip terminal swing moment (Nm/kg) Hip pre−swing moment (Nm/kg) Knee pre−swing moment (Nm/kg) Ankle pre-swing moment (Nm/kg) Ankle terminal joint moment (Nm/kg) Time to maximum pronation (% stance) Calcaneus to vertical touchdown angle (°) Ankle dorsiflexion angle (°)
26.3 (3.9) 5.46 (3.61) 5.92 (3.95) 0.82 (5.73) 0.06 (3.21) 14 (5) −3 (4) 41 (4) 7.3 (4.8) –0.80 (0.34) 11 (3) 330 (59) 41 (7) −0.01 (0.24) −0.80 (0.64) −0.37 (0.21) 0.10 (0.09) −0.01 (0.07) −0.03 (0.02) 40.32 (1.38) −6.97 (0.73) Effect Size (d) 0.86 1.08 1.03 0.89
22.0 (5.5) 1.17 (3.78) 2.64 (4.74) −2.94 (6.18) 0.82 (3.05) 9 (3) −5 (3) 37 (7) 3.1 (3.8) –0.33 (0.35) 13 (3) 300 (39) 45 (7) 0.33 (0.32) 0.21 (0.70) −0.10 (0.25) 0.01 (0.08) 0.08 (0.18) 0.00 (0.01) 37.30 (2.31) −8.94 (1.54)
−0.90 −1.16 −0.75 −0.63 0.24 −1.21 −0.57 −0.70 −0.97 1.36 0.67 −0.62 0.57 1.20 1.51 1.17 −1.06 0.66 1.90 −1.72 −1.82
−1.52, −0.25* −2.01, −0.22* −1.59, 0.14 −1.46, 0.25 −0.61, 1.07 −2.11, −0.21* −1.43,0.35 −1.57, 0.23 −1.95, 0.11 0.21, 2.36* 0.07, 1.24* −1.19, −0.02* −0.02, 1.14 0.51, 1.85* 0.78, 2.18* 0.48, 1.81* −1.70, −0.38* 0.01, 1.28* 1.12, 2.60* −2.21, −1.20* −2.32, −1.30* P-Value 0.033* 0.01* 0.062* 0.029*
Hein et al. [31]
Williams III et al. [29] Ryan et al. [30]
Kim et al. [26]
McCrory et al. [28] Donoghue et al. [24]
Ankle eversion angle (°) Calcaneal angle (°) Knee flexion angle (stance phase) (°) (Mean(SD)/CI data not provided by Donoghue et al. [24]). * Statistically significant.
impact on all other aspects of gait biomechanics, resulting in uneven ground reaction forces and plantar pressures, less optimum joint positions and angles and reductions in temporospatial parameters. Addressing this altered neuromotor control in individuals with Achilles tendinopathy should therefore form a more significant aspect of the rehabalitation process.
(d = 1.33) and earlier offset of muscle activity (d = 1.14) in the Gluteus Maximus of subjects with Achilles tendinopathy [25]. This is the only study to date to demonstrate significant differences in muscle activity of Gluteus Maximus in participants with Achilles tendinopathy. Reduced muscle activity of Gluteus Maximus (primary hip extensor during gait) and subsequent reduced hip joint moment has been reported to increase joint moment further along the kinetic chain [34] and thus could be a possible mechanism for the development of Achilles tendinopathy. Baur et al. [22] reported no significant differences between the two groups in muscle activity of Tibialis anterior, however reported reduced muscle activity in the peroneus longus (d = −0.54) and medial gastrocnemius muscles (d = −0.63) during the early stance phase in subjects with Achilles tendinopathy [22]. Wyndow et al. reported an earlier relative offset timing between Soleus and lateral gastrocnemius (d = 0.90) in Achilles tendinopathy subjects, however no significant differences in muscle onset timing [27]. Reductions in muscle activity and delay in muscle timing is likely to have a profound
5.1.4. Kinetics McCrory et al. is the only study to date to investigate timing of ground reaction forces and reported a number of significant differences between individuals with Achilles tendinopathy and health controls. Significant increases in individuals with Achilles tendinopathy were noted for time to first vertical peak (d = 12.9), time to minimum vertical peak (d = 18.4), time to max medial force (d = 13.25). Time to second vertical force (d = −19.50), time to max propulsive force (d = −26.49), time to max braking force (d = −56.12), time to max lateral force (d = −12.05) and average propulsive force (d = −9.38)
Fig. 8. Kinematic variables Effect Sizes.
154
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
5.2. Limitations
were all significantly reduced in individuals with Achilles tendinopathy.[28] As participants were symptomatic during this study, the delay in reaching peak forces and increase in braking forces may likely represent a protective mechanism to minimise further stress to the Achilles tendon.
Limitations exist in the study quality described earlier, particularly study design in which only 3 out of 14 studies utilised a prospective design increasing the difficulty of differentiating between cause and effect. All reported studies were carried out in active people, despite a large cohort study indicating the majority of Achilles tendinopathy is not related to sports activity [8]. Tendinopathy type was not specified in half of the studies making it difficult to understand and establish pathophysiology. The majority of the sample sizes of the included studies were relatively small, possibly distorting effect size calculations between groups. Another important limitation is that the biomechanical variables examined vary significantly between studies making it difficult to make comparisons and thus draw conclusions.
5.1.5. Kinematics Donoghue et al. (1) reported a reduced leg abduction max angle (angle between the lower leg and the ground on the medial side as viewed from the posterior-indicating level of varus/valgus of lower leg) in barefoot individuals with Achilles tendinopathy (d = −1.16) however no significant difference was noted when shod [24]. This reduction in leg abduction correlates with the significant reduction in onset and duration of activation of the Gluteus medius in individuals with Achilles tendinopathy reported by Franettovich et al. [25].
5.3. What this review adds 5.1.5.1. Foot kinematics. 2 out of the 6 studies investigating foot kinematics reported significant differences between healthy individuals and individuals with Achilles tendinopathy. McCrory et al. reporting a significant increase in time to maximum pronation (d = −1.72) and greater calcaneal inversion (increase in calcaneus to vertical touchdown angle) (d = −1.82).[28] Donoghue et al. (2) also reported greater calcaneal inversion at heel-strike (d = 1.03).[23] These findings are in keeping with Clement et al. widely cited excessive pronation following an inverted heel-strike hypothesis [35].
This is the second review of lower-limb gait biomechanics in individuals with Achilles tendinopathy. This study includes data from 5 new studies since the last review [18]. This update identifies and calculates effect sizes for eighteen new significant gait biomechanical characteristics in individuals with Achilles tendinopathy.
5.4. Clinical implications Critical appraisal of the new biomechanical characteristics identified have highlighted the importance of the potential role of gluteal neuromuscular control rehabilitation as a management strategy in Achilles tendinopathy. Important differences in temporospatial characteristics between healthy individuals and those with Achilles tendinopathy require further investigation to quantify the association precisely enabling useful applications of these characteristics including during screening and diagnosis and also as return to play markers during rehabilitation.
5.1.5.2. Ankle kinematics. Donoghue et al. (2) reported increased ankle dorsiflexion in Achilles tendinopathy (d = 0.86) [23]. Contrastingly Hein et al. reported a lower maximum ankle dorsiflexion angle (d = −1.21) [31] and Ryan et al. reported a reduced maximum ankle dorsiflexion velocity (d = 0.62) [30]. Notably Hein et al. is the only study investigating gait kinematics in Achilles tendinopathy with a prospective study design. Donoghue et al. (2) and Ryan et al. both reported a greater ankle eversion angle (d = 1.08) and (d = 0.67) respectively [23,30] and Kim et al. reported an increase in ankle joint moment during the pre-swing phase (d = 0.66) and the terminal swing phase in individuals with Achilles tendinopathy (d = 1.90) [26]. These findings of increased ankle eversion are also in keeping with the excessive pronation ‘whipping action’ hypothesis cited by Clement et al. drawing the Achilles tendon medially leading to microtrauma [35].
6. Conclusion Eighteen new biomechanical characteristics were established in individuals with Achilles tendinopathy, many which have significant potential clinical implications in the prevention and management of Achilles tendinopathy. Runners with Achilles tendinopathy have been reported to have reductions in temporospatial characteristics including stride length, step length and gait speed. Individuals with Achilles tendinopathy have been shown to have increased ankle eversion, ankle and hip joint moments, time to maximum pronation, calcaneal inversion and a decrease in pre-swing phase knee flexion. Individuals with Achilles tendinopathy were also reported to have reductions in muscle activity of tibialis anterior, rectus femoris, peroneus longus, medial gastrocnemius and the gluteal maximus and medius muscles. A reduction in onset of gluteal maximus and medius muscle activity and an increase in offset of soleus and lateral gastrocnemius muscles were also identified. Additionally, altered plantar pressures and differences in the timing of ground reaction forces were established. Clinically we have established the need to emphasise neuromuscular control during the management and rehabilitation process in individuals with Achilles tendinopathy. Due to weak study designs, underpowered studies and inconsistency in reported variables, interpretation of findings must be done with caution. Future studies should focus on high powered prospective studies utilising consistent variables enabling aetiology to be established and subsequently the implementation of preventive and therapeutic measures.
5.1.5.3. Knee kinematics. Azevedo et al. reported reduced knee flexion between heel strike and midstance (d = −0.90) [10], whereas Donoghue et al. (2) reported increased knee flexion (d = 0.89) throughout the entire stance phase in runners with Achilles tendinopathy [23].This correlates with the increased levels of knee flexion observed in Achilles tendinopathy subjects with excessive pronation [36]. Kim et al. reported a significant decrease in the preswing knee moment phase (d = −1.06) in runners with Achilles tendinopathy [26]. This correlates with the reduced hip extensor activity reported by Franettovich et al. resulting in increased forces distally at the Achilles tendon [25]. 5.1.5.4. Hip kinematics. Data on hip joint kinematics is very limited with only data from 2 studies existing. Kim et al. reported a significant increase in hip joint moment for mid-stance (d = 1.20), terminal stance (d = 1.51) and pre-swing phases (d = 1.17) in runners with Achilles tendinopathy [26]. This is in contrast to EMG data by Franettovich et al. reporting reduced hip extensor activity in individuals with Achilles tendinopathy [25]. Azevedo et al. reported no significant differences between healthy individuals and runners with Achilles tendinopathy [10]. 155
Gait & Posture 62 (2018) 146–156
I. Ogbonmwan et al.
Funding
[17] J. Perry, J.M. Burnfield, Gait Analysis: Normal and Pathological Function, Second Edition. Gait Analysis: Normal and Pathological Function, 2nd ed., Slack Inc, Thorofare, 2010. [18] S.E. Munteanu, C.J. Barton, Lower limb biomechanics during running in individuals with achilles tendinopathy: a systematic review, J. Foot Ankle Res. 4 (2011) 15. [19] A. Liberati, D.G. Altman, J. Tetzlaff, C. Mulrow, P.C. Gotzsche, J.P. Ioannidis, et al., The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration, J. Clin. Epidemiol. 62 (10) (2009) e1–34. [20] J.P.T. Higgins, D.G. Altman, P.C. Gotzsche, P. Juni, D. Moher, A.D. Oxman, et al., The Cochrane Collaboration's tool for assessing risk of bias in randomised trials, Br. Med. J. 343 (2011) 9. [21] H. Baur, C. Divert, A. Hirschmüller, S. Müller, A. Belli, F. Mayer, Analysis of gait differences in healthy runners and runners with chronic Achilles tendon complaints, Isokinet. Exerc. Sci. 12 (2) (2004) 111–116. [22] H. Baur, S. Muller, A. Hirschmuller, M. Cassel, J. Weber, F. Mayer, Comparison in lower leg neuromuscular activity between runners with unilateral mid-portion Achilles tendinopathy and healthy individuals, J. Electromyogr. Kinesiol. 21 (3) (2011) 499–505. [23] O.A. Donoghue, A.J. Harrison, N. Coffey, K. Hayes, Functional data analysis of running kinematics in chronic Achilles tendon injury, Med. Sci. Sports Exerc. 40 (7) (2008) 1323–1335. [24] O.A. Donoghue, A.J. Harrison, P. Laxton, R.K. Jones, Lower limb kinematics of subjects with chronic achilles tendon injury during running, Res. Sports Med. 16 (1) (2008) 23–38. [25] M.M. Franettovich Smith, C. Honeywill, N. Wyndow, K.M. Crossley, M.W. Creaby, Neuromotor control of gluteal muscles in runners with achilles tendinopathy, Med. Sci. Sports Exerc. 46 (3) (2014) 594–599. [26] S. Kim, J. Yu, Changes of gait parameters and lower limb dynamics in recreational runners with achilles tendinopathy, J. Sports Sci. Med. (2015) 1303–2968. [27] N. Wyndow, S.M. Cowan, T.V. Wrigley, K.M. Crossley, Triceps surae activation is altered in male runners with Achilles tendinopathy, J. Electromyogr. Kinesiol. 23 (1) (2013) 166–172. [28] J.L. McCrory, D.F. Martin, R.B. Lowery, D.W. Cannon, W.W. Curl, H.M. Read Jr.et al., Etiologic factors associated with Achilles tendinitis in runners, Med. Sci. Sports Exerc. 31 (10) (1999) 1374–1381. [29] I.D.S.B. Williams, J.A. Zambardino, V.A. Banning, Transverse-plane mechanics at the knee and tibia in runners with and without a history of achilles tendonopathy, J. Orthop. Sports Phys. Ther. 38 (12) (2008) 761–767. [30] M. Ryan, S. Grau, I. Krauss, C. Maiwald, J. Taunton, T. Horstmann, Kinematic analysis of runners with achilles mid-Portion tendinopathy, Foot Ankle Int. 30 (12) (2009) 1190–1195. [31] T. Hein, P. Janssen, U. Wagner-Fritz, G. Haupt, S. Grau, Prospective analysis of intrinsic and extrinsic risk factors on the development of Achilles tendon pain in runners, Scand. J. Med. Sci. Sports 24 (3) (2014) e201-121p. [32] K.R. Kaufman, S.K. Brodine, R.A. Shaffer, C.W. Johnson, T.R. Cullison, The effect of foot structure and range of motion on musculoskeletal overuse injuries, Am. J. Sports Med. 27 (5) (1999) 585–593. [33] A. Van Ginckel, Y. Thijs, N.G. Hesar, N. Mahieu, D. De Clercq, P. Roosen, et al., Intrinsic gait-related risk factors for Achilles tendinopathy in novice runners: a prospective study, Gait Posture 29 (3) (2009) 387–391. [34] H. Sadeghi, S. Sadeghi, F. Prince, P. Allard, H. Labelle, C.L. Vaughan, Functional roles of ankle and hip sagittal muscle moments in able-bodied gait, Clin. Biomech. 16 (8) (2001) 688–695. [35] D.B. Clement, J.E. Taunton, G.W. Smart, Achilles tendinitis and peritendinitis: etiology and treatment, Am. J. Sports Med. 12 (3) (1984) 179–184. [36] I. McClay, K. Manal, A comparison of three-dimensional lower extremity kinematics during running between excessive pronators and normals, Clin. Biomech. 13 (3) (1998) 195–203.
None obtained. Conflicting interests None to be declared. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at https://doi.org/10.1016/j.gaitpost.2018.03.010. References [1] N. Maffulli, K.M. Khan, G. Puddu, Overuse tendon conditions: time to change a confusing terminology, Arthroscopy: J. Arthrosc. Relat. Surg. 14 (8) (1998) 840–843. [2] N. Maffulli, D. Kader, Tendinopathy of tendo achillis, J. Bone Joint Surg. [Br] 84 (1) (2002) 1–8. [3] M. Paavola, P. Kannus, T.A.H. Jarvinen, K. Khan, L. Jozsa, M. Jarvinen, Current concepts review achilles tendinopathy, J. Bone Joint Surg. [Am] 84A (11) (2002) 2062–2076. [4] X.F. Courville, M.P. Coe, P.J. Hecht, Current concepts review: noninsertional Achilles tendinopathy, Foot Ankle Int. 30 (11) (2009) 1132–1142. [5] M. Kvist, Achilles tendon injuries in athletes, Sports Med. 18 (3) (1994) 173–201. [6] F.E. Johannsen, A.N. Gam, [Achillodynia is not just a sports injury], Ugeskr. Laeger 172 (48) (2010) 3325–3329. [7] J. Lysholm, J. Wiklander, Injuries in runners, Am. J. Sports Med. 15 (2) (1987) 168–171. [8] S. de Jonge, C. van den Berg, R.J. de Vos, H.J. van der Heide, A. Weir, J.A. Verhaar, et al., Incidence of midportion Achilles tendinopathy in the general population, Br. J. Sports Med. 45 (13) (2011) 1026–1028. [9] A.V. Lorimer, P.A. Hume, Achilles tendon injury risk factors associated with running, Sports Med. 44 (10) (2014) 1459–1472. [10] L.B. Azevedo, M.I. Lambert, C.L. Vaughan, C.M. O’Connor, M.P. Schwellnus, Biomechanical variables associated with Achilles tendinopathy in runners, Br. J. Sports Med. 43 (4) (2009) 288–292. [11] N. Wyndow, S.M. Cowan, T.V. Wrigley, K.M. Crossley, Neuromotor control of the lower limb in Achilles tendinopathy: implications for foot orthotic therapy, Sports Med. 40 (9) (2010) 715–727. [12] H.K. Vincent, D.C. Herman, L. Lear-Barnes, R. Barnes, C. Chen, S. Greenberg, et al., Setting standards for medically-based running analysis, Curr. Sports Med. Rep. 13 (4) (2014) 275–283. [13] E. Ceseracciu, Z. Sawacha, C. Cobelli, Comparison of markerless and marker-based motion capture technologies through simultaneous data collection during gait: proof of concept, PLoS One 9 (3) (2014) e87640. [14] F. Dierick, M. Penta, D. Renaut, C. Detrembleur, A force measuring treadmill in clinical gait analysis, Gait Posture 20 (3) (2004) 299–303. [15] P.O. Riley, G. Paolini, U. Della Croce, K.W. Paylo, D.C. Kerrigan, A kinematic and kinetic comparison of overground and treadmill walking in healthy subjects, Gait Posture 26 (1) (2007) 17–24. [16] J.J. Rechtien, J.B. Gelblum, A.J. Haig, A.J. Gitter, Technology assessment: dynamic electromyography in gait and motion analysis, Muscle Nerve 19 (3) (1996) 396–402.
156