Hip joint mechanics during walking in individuals with mild-to-moderate hip osteoarthritis

Accepted Manuscript Title: Hip joint mechanics during walking in individuals with mild-to-moderate hip osteoarthritis Authors: Maria Constantinou, Aderson Loureiro, Christopher Carty, Peter Mills, Rod Barrett PII: DOI: Reference:

S0966-6362(17)30026-7 http://dx.doi.org/doi:10.1016/j.gaitpost.2017.01.017 GAIPOS 5296

To appear in:

Gait & Posture

Received date: Revised date: Accepted date:

18-12-2015 20-12-2016 23-1-2017

Please cite this article as: Constantinou Maria, Loureiro Aderson, Carty Christopher, Mills Peter, Barrett Rod.Hip joint mechanics during walking in individuals with mild-to-moderate hip osteoarthritis.Gait and Posture http://dx.doi.org/10.1016/j.gaitpost.2017.01.017 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Original Article Title: Hip joint mechanics during walking in individuals with mild-to-moderate hip osteoarthritis Author names and affiliations: Maria Constantinou1,2, PhD, Aderson Loureiro1, PhD, Christopher Carty1,3, PhD, Peter Mills1, PhD, Rod Barrett1, PhD 1

School of Allied Health Sciences and Menzies Health Institute Queensland, Griffith

University, Gold Coast campus, 4222, Queensland, Australia 2

School of Physiotherapy, Faculty of Health Sciences, Australian Catholic University, Banyo,

Brisbane, Australia 3

Children's Health Queensland, Brisbane, Australia

Word count: 3723 words; Tables: 3, Figures: 2 (Colour, as Supplementary material) Corresponding author: Maria Constantinou School of Physiotherapy, Faculty of Health Sciences, Australian Catholic University, Brisbane, QLD 4014, Australia Tel: 07 3623 7590

Fax: 07 5552 8674

Email: [email protected]; [email protected] Co-authors email addresses: [email protected];

[email protected];

[email protected];

[email protected]

1

Highlights 

Individuals with mild-to-moderate hip osteoarthritis (OA) walk slower



The hip experiences less joint range and lower moments in mild-to-moderate hip OA



Altered hip joint mechanics may impact on hip tissue mechano-biology in hip OA

Abstract The purpose of this case-control study was to characterise hip joint kinematics and moments during gait in people with mild-to-moderate hip osteoarthritis (OA). Eligible participants were allocated to the hip OA group (n = 27) or the age-matched control group (n = 26) based on radiographic and symptomatically defined inclusion criteria. Participants walked barefoot along a 10-m walkway at their self-selected gait speed. Trajectories of 43 markers attached to the trunk, pelvis, upper and lower limbs were recorded using a 12-camera motion capture system. Ground reaction force data were simultaneously collected. Individuals in the hip OA group had a 10% higher body mass, 13% slower self-selected walking speed, 10% shorter step length, 2% and 9% longer relative stance and double support duration (% stride) respectively, 41% lower sagittal plane hip range of motion, and 28% and 45% lower peak sagittal and transverse plane hip moments respectively during gait compared to controls (p < 0.05). The finding that individuals with mild-to-moderate hip OA experienced less net hip joint loading over a reduced range of hip motion for a longer proportion of the gait cycle when walking at their preferred gait speed suggest that the mechanics of the hip joint are altered in hip OA, and could have implications for disease progression through altered mechano-biological processes within the joint. Keywords: hip osteoarthritis; mild-to-moderate; gait; kinetic; kinematic; hip joint mechanics.

2

1. Introduction Hip joint osteoarthritis (OA) is a chronic and commonly progressive musculoskeletal condition that results in pain, functional limitations [1] and significantly reduced quality of life [2]. Walking in particular is an important and frequent activity of daily living that is commonly affected by hip OA [3]. Understanding how gait mechanics are altered in hip OA is important for several reasons. Firstly, gait analysis provides an objective way to characterize the functional limitations at different stages of the disease which could be used to inform the development and timely use of targeted therapeutic interventions such as exercise. Secondly, gait analysis can provide important information about the mechanics of the hip joint, which could help identify mechanical risk factors associated with development and/or progression of the disease. Although OA has been regarded as a disease caused primarily by excessive wear and tear, a more recent hypothesis is that joint degeneration occurs when joint tissues experience loading which is incompatible with the tissue state [4]. Within this framework nonoptimal loading including over-loading and/or under-loading of joint structures could contribute to joint degeneration [4]. Indeed less peak hip joint excursion and lower external hip joint moments during gait have been associated with disease severity in advanced stage hip OA [5, 6] while less sagittal plane hip joint excursion during gait has been linked to greater size of cartilage lesions in mild-to-moderate radiographic hip OA [7]. Further studies that better characterize joint level mechanics in hip OA appear to have the potential to improve

3

understanding of the mechanical factors that contribute to disease onset and progression and would also inform mechano-biological models of OA [8]. Most gait studies in individuals with hip OA to date have focused on the advanced stage of the disease [5, 9-11] or have evaluated groups of mixed disease severity [6] relative to healthy controls. From these studies it is clear that relative to healthy controls, individuals with advanced stage hip OA have a slower self-selected walking speed [3], shorter step and stride length and longer stance duration [3], less hip excursion in the sagittal plane [9, 10] and lower hip joint moments in the sagittal and frontal planes during the stance phase of gait [9-11]. Individuals with greater disease severity have also been reported to exhibit more pronounced gait deficits in hip and pelvis kinematics with faster or slower walking speeds than their selfselected gait speed [12]. There is also evidence that gait mechanics are altered in individuals with mild-to-moderate stage hip OA relative to healthy, age-matched controls. These differences include 7–12% slower self-selected walking speed [13, 14], shorter stride length [14], less hip joint extension [7, 14, 15], hip abduction [14, 15] and sagittal hip and knee excursions [13], greater hip flexion range [7] and lower peak external hip flexion [13, 14] and abduction moments [14] during stance in mild-moderate hip OA. Of these studies, two compared kinematics and/or kinetics in hip OA versus controls at a fixed gait speed [7, 15] and two reported joint moments and/or powers normalized to body mass for walking at a self-selected gait speed [13, 14]. Though selfselected gait speed is reduced in hip OA [3], which might be expected to unload the joint, and body weight tends to be higher in hip OA [16], which could alone account for lower normalized joint moments and powers in those with hip OA relative to controls. A distinction between absolute and normalized joint moments is important from a mechano-biological perspective, as absolute joint moments are likely to be a better proxy for the load experienced by the joint than normalized moments. At present, however, information on the independent effects of 4

slower self-selected walking speed and greater body mass on absolute joint moments during gait, in individuals with mild-to-moderate hip OA relative to healthy controls, is lacking. The purpose of this study was to characterise hip joint kinematics and absolute hip joint moments during self-selected walking speed in people with mild-to-moderate hip OA and healthy, age-matched controls. It was hypothesised that individuals with mild-to-moderate hip OA would experience differences in magnitude and timing of hip joint kinematic and/or kinetic measures relative to healthy age-matched controls. The extent to which any group differences in selected joint level mechanics were driven by group differences in gait speed and or body mass were also investigated. 2. Methods 2.1 Participants Participants over the age of 45 years were recruited through advertising, word of mouth and from hospital Orthopaedic waiting lists from June 2011 to October 2014. Volunteers were screened using the self-reported modified Harris Hip Score (HHS) [17] and radiographic examination. Anterior-posterior radiographs of the pelvis and hips (bilateral) were undertaken in weight-bearing, with feet internally rotated by 15 ± 5° [18] on a custom built platform, with foot map for standardisation of the feet position [19]. The film distance to the source was set at 100 cm with the X-ray beam horizontal [18]. An experienced radiologist electronically measured the supero-medial, apical and supero-lateral hip joint space width (JSW) and scored the hip radiographs for both hips based on the Kellgren-Lawrence (K-L) grades, and the presence of osteophytes [20, 21] using the OARSI grading criteria [22]. Volunteers with K-L grade 2 or 3 [23] and/or JSW ≤ 3 mm in one or both hips, experiencing hip pain in the last 3 months and HHS ≤ 95 points were allocated to the hip OA group and volunteers with K-L grades ≤ 1 and JSW > 3 mm in both hips, absence of any hip pain or symptoms and HHS > 95

5

points were allocated to the control group. Individuals were excluded if they reported any major lower limb musculoskeletal or neurological conditions besides hip OA. Ethical approval was granted by the Human Research Ethics Committees from the relevant institutions and all participants signed an informed consent prior to commencement of the study. A priori power calculation (G*Power 3 [24] ) for the sample size required to detect 11.08° mean difference in hip extension ROM (Hip OA = 1.77 ± 10.3°, Controls = -9.31 ± 6.27°, Effect Size = 1.34) based on a previous case-control study by Watelain et al (2001) [14], estimated minimum 8 participants required in each group (significance level was set at α = 0.05 and power at 0.80 (one tail), with allocation ratio N=1). From 420 individuals who volunteered, 60 individuals with hip pain and 48 with no symptoms were eligible to complete the HHS and subsequently undergo radiographic screening evaluation for possible participation in the study. In total 27 participants in the hip OA group and 26 in the control group, all were of Caucasian background, were eligible and were included in the study. The hip OA group consisted of 11 individuals with bilateral hip OA and 16 participants with unilateral hip OA. As there were no significant differences in any of the demographic or gait measures between these sub-groups data were pooled for subsequent analyses. Data for the affected unilateral or most affected bilateral limb of individuals with hip OA and for a randomly selected index limb for the controls were used for analysis. 2.2 Patient-reported outcomes Patient-reported outcomes were evaluated using the modified HHS [17]. The HHS is a selfadministered questionnaire that incorporates one question with up to 44 points based on pain and 6 questions with up to 47 points based on function, totalling 91 points, which are multiplied by a factor of 1.1 to give a score out of 100 points [25]. Higher scores refer to more function. 6

2.3 Gait protocol Gait analysis was performed with participants walking barefoot at self-selected speeds over a 10-meter walkway. Following several familiarisation trials, participants were asked to walk at a “comfortable pace, as you would normally walk without feeling unsafe”. A total of 3-5 successful trials were recorded for each participant. A walking trial was deemed successful when the participant stepped on each of the two force platforms with alternate feet.

2.4 Gait data collection and analysis procedures A 12-camera VICON motion analysis system (VICON, Oxford Metrics, Oxford, UK) operating at 200 Hz was used to record three-dimensional (3D) marker trajectories according to the University of Western Australia (UWA) full body gait model [26], which consisted of 43 markers attached to specific anatomical landmarks for the lower limb, which included a cluster of four markers attached to each of the shank and thigh segments. All markers were attached using double-sided tape. Ground reaction force data was simultaneously collected at 2000 Hz with two 900 x 600 mm piezoelectric force platforms (Type 9287A, Kistler Instrument Corporation, USA) embedded in the centre of the 10-metre walkway. Marker coordinate trajectories and force plate data were filtered using a fourth order zero lag Butterworth Filter, with a cut-off frequency of 6 Hz. Filtered marker trajectories were used to compute 3D segment (trunk and pelvis) and joint (hip, knee and ankle) kinematics using BodyBuilder modelling software, version 3.6 (Vicon; Oxford Metrics). The hip joint centre was estimated using the Harrington equation [27] in accordance with Kainz et al. (2015)[28], the knee joint centre and axis was defined using the medial and lateral femoral epicondyle markers and the ankle joint centre and axis was defined using the medial and lateral malleoli markers [26]. Segment and joint angles were calculated using the Euler angle method in a flexion/extension, abduction/adduction, and internal/external rotation

7

sequence. Joint moments were computed using inverse dynamic equations, incorporating a mathematical link-segment model of the limb using body segment parameters reported by de Leva (1996) [29] and working back from the kinematics to derive the kinetics responsible for the motion [26, 30]. Spatial-temporal, kinematic and kinetic data for individual gait cycles were extracted and collated using the Biomechanical ToolKit (BTK) [31] and custom software developed in Matlab (Release 2015a, Mathworks, Natick, MA). Joint angle and joint moment data were time normalised to 100% of a gait cycle. Internal joint moments were reported in absolute units (Nm). Data were graphically viewed for each trial to ensure all included gait trials were valid. Joint excursion angles for the pelvis and hip were calculated by subtracting the minimum joint range from the maximum during the stance phase. The kinematic measures of interest were peak anterior pelvic tilt and pelvic superior/inferior list (defined as pelvic obliquity) during stance and peak joint range for hip flexion, extension and adduction, knee flexion and extension and ankle dorsi and plantar-flexion angles during stance. The kinetic measures of interest were peak hip flexion and extension internal joint moments during stance, peak hip abduction moments during mid-stance and during early and late stance (to compare the two peaks occurring during stance at a similar time point for each group), peak knee extension and flexion moments and peak plantar-flexion moments during stance. 2.5 Statistical analysis Data was inspected for normal distribution using Shapiro-Wilk test in the first instance and further explored for outliers using box-plots. For the ordinal scale of K-L scores the median and interquartile range were calculated for each group. A univariate General Linear Model was used to identify the effect of group (hip OA versus control) on the demographic, spatial-

8

temporal, kinematic and kinetic variables. This analysis was repeated with self-selected walking speed as a co-variate in the General Linear Model, to determine which group differences persisted after controlling for gait speed. For joint moments, the analysis was also repeated with body mass as a covariate. All statistical analysis was performed using the Statistical Package for the Social Sciences (SPSS, Version 22, IBM Corporation). Significance was accepted for P <0.05. 3. Results 3.1 Participant Characteristics Individuals with hip OA had a significantly higher mass and Body Mass Index (BMI), and lower joint space width, higher Kellgren-Lawrence score and lower Harris Hip Score than the control group (Table 1). The median K-L score for the index limb of the hip OA group was 3.0 (IQR = 1), and for the index limb of the control group 0.0 (IQR = 1). 3.2 Spatial-temporal gait characteristics Individuals with hip OA walked significantly slower with shorter step length than healthy controls. Stance and double support duration were significantly longer in the hip OA group compared to healthy controls. Only the relative stance and swing duration (% gait cycle) remained significantly lower in the hip OA group compared to the controls after accounting for group differences in gait speed. Spatial-temporal gait characteristics for each group are summarised in Table 2. 3.3 Kinematic gait characteristics

9

Ensemble averages for pelvis, hip, knee and ankle kinematics are depicted in Supplementary Figure 1. Individuals with hip OA exhibited significantly lower peak pelvic obliquity in early stance and lower peak hip extension during the late stance phase compared to healthy controls. The key kinematic findings at selected time points are described in Table 3. There was significantly less overall joint excursion for pelvis obliquity (hip OA: 5.4 ± 2.7°, control: 7.5 ± 2.5°; F=8.5, p=0.005), hip flexion/extension (hip OA: 37.5 ± 7.2°, control: 43.8 ± 4.7°; F=14.0, p<0.01) and hip abduction/adduction (hip OA: 9.3 ± 2.7°, control: 10.8 ± 2.1°; F=5.7, p=0.021) angles in individuals with hip OA compared to healthy controls during stance. 3.4 Internal hip, knee and ankle joint moments Ensemble averages for 3D internal hip joint moments are depicted in Supplementary Figure 2. The peak hip flexion moment at late stance (toe-off) and the peak external rotation moment at early stance (foot strike) were significantly lower, while at mid-stance the minimum abduction moment was significantly higher in those with hip OA relative to controls (Table 3). After accounting for self-selected gait speed and body mass separately, peak internal rotation moments at foot strike remained significantly lower in those with hip OA compared to healthy controls (p < 0.05). After accounting for body mass, peak hip flexion moment at toe-off, peak hip extension and flexion moments during stance and peak hip abduction moments during late stance were significantly lower (p < 0.05) in those with hip OA relative to controls. 4. Discussion This study demonstrated that individuals with mild-to-moderate hip OA walked with a slower self-selected gait speed, shorter step length, longer stance duration, less hip range of motion and experienced lower hip sagittal and transverse absolute joint moments during the stance 10

phase compared to the control group. Overall these findings suggest that, despite individuals in the hip OA group having a higher body mass, the mechanics of the joint are altered in mildto-moderate hip OA in a manner consistent with under-loading. The region of cartilage exposed to loading during walking at self-selected gait speed may also be reduced in hip OA as a result of less hip extension during stance. The observed average slower self-selected gait speed of 13% in individuals with hip OA in the present study is in relative agreement with the 7% and 12% slower gait speeds reported respectively by Watelain et al. [14] and Eitzen et al. [13] in mild-to-moderate hip OA. In a recent systematic review and meta-analysis of spatial-temporal gait characteristics in individuals with hip OA an average gait speed reduction of 26% was reported [3]. Given that most included participants in this meta-analysis were in the advanced stages of hip OA, it appears that self-selected gait speed declines with disease progression. The slower gait speed in the hip OA group in the current study was due to shorter step length on the affected limb in individuals with hip OA relative to controls, and together with the observed longer relative double support and stance duration (% stride) in the affected limb, it may likely be caused by a combination of neuromuscular deficits, metabolic factors and/or pain. Furthermore, the longer relative stance duration (% stride) may be used as a strategy of slowing gait speed, as it may have the added benefit of distributing joint loads over a longer period during stance. Selfselected gait speed was found to be the main driver of group differences in spatial-temporal parameters with only relative stance duration (% stride) remaining different after controlling for gait speed. Findings of significantly less sagittal hip range during self-selected speed gait, with 41% less hip extension in individuals with hip OA relative to controls, are consistent with previous studies on individuals with mild-to-moderate hip OA [13, 14, 32] and were explained particularly by a decrease in hip extension during late stance. Similar to findings of Watelain 11

et al. [14], we also found that individuals in the hip OA group experienced less pelvic obliquity in the frontal plane than healthy controls during early stance. Although the lesser sagittal hip range of motion in individuals with hip OA did not persist after accounting for group differences in self-selected walking speed, they are nevertheless likely to decrease the surface area of hip cartilage being loaded during the stance phase of natural gait. Focal loading of the hip cartilage region within the available range of motion together with longer stance phase and the relative unloading in the region beyond the limits of the hip joint range of motion may therefore be factors that contribute to disease progression [4]. Indeed, a recent modelling study demonstrated that peak cartilage stresses are lowered in an arthritic hip joint when loading patterns associated with pathological gait were applied [33]. The current study demonstrated that the slower walking speed is more influential than higher body mass on absolute hip joint moments during self-selected walking speed in individuals with mild-to-moderate hip OA compared to healthy controls. Although body mass was on average 10% higher in the hip OA group and might be expected to contribute to higher hip moments during walking than in healthy controls, peak internal hip joint moments during stance in the sagittal and transverse planes were 28 and 45% lower in the hip OA group respectively. The finding of lower peak hip flexion moment at toe-off is consistent with previous gait studies in individuals with mild-to-moderate hip OA that reported lower peak hip flexion moments at toe-off [14] and lower peak hip flexion and extension moments in those with advanced stages of hip OA relative to healthy controls [6, 9], and suggest that the symptomatic hip OA limb experiences a general under-loading in the sagittal plane when walking at a self-selected gait speed. While both groups had similar absolute joint moments in the frontal plane at early and late stance, at mid-stance individuals in the hip OA group had 18% higher internal hip abduction moments than the controls. This finding is indicative of an inability to modulate hip abduction moments during mid-stance and may reflect a degree of 12

reduced neuromuscular control in the hip OA group. The fact that the peak hip abduction moment in late stance was significantly lower in those with hip OA after accounting for body mass, suggests that hip abductor function may be compromised in hip OA, and is consistent with hip abductor muscle weakness reported in advanced stages of the disease [11]. Exercise interventions that target the deficits in gait identified in this study, with a focus on hip range and hip muscle strength, performed within the constraints imposed by pain could be efficacious in improving gait function in this group. After adjusting for the effect of body mass, group differences in hip moments became more apparent due to the higher body mass in the hip OA group. Although this finding does not alter the general conclusion of our study, it does suggest normalising joint moments to body mass may provide different information in studies comparing groups with dissimilar body mass; group differences in normalised joint moments could be driven by differences in body mass in the absence of differences in absolute raw joint moments. Conversely, after adjusting for selfselected gait speed, only evidence of lower peak transverse plane hip moments remained for the hip OA group. Despite this finding that joint moments in hip OA were to some degree explained by a slower self-selected gait speed, the lower hip moments in the hip OA group relative to controls when walking at preferred gait speed remain of practical significance from a tissue level perspective, as these loads are the loads the joint experiences in natural gait. The effect of this altered hip joint loading in hip OA might be expected to alter the mechanobiology of the joint and have possible implications for disease progression [4, 8]. Several limitations of this study need to be acknowledged. The participants with bilateral and unilateral hip OA were grouped in one cohort and even though no differences were identified in the key variables, this could be a confounding factor, as individuals with bilateral hip OA may not be able to compensate with the contra-lateral limb in the same way as individuals with unilateral hip OA [34]. It would thus be beneficial in future studies to compare gait 13

characteristics in individuals with bilateral and with unilateral hip OA and to investigate the symmetry of gait in these two groups as well as their preferred strategies for altering their gait speed. As this was a cross-sectional study, it is unclear whether the gait characteristics identified in this study were present before, or emerged after, the onset of the disease. The extent to which the observed gait characteristics in mild-to-moderate hip OA are related to factors such as pain, lower limb muscle weakness, or reduced passive joint range of motion also remain unknown. Net hip joint moments obtained using inverse dynamics were used in the present study as a proxy measure of hip joint loading. In future, it will be of benefit to use more sophisticated measures of articular loading that better reflect subject-specific musculoskeletal geometry and muscle activation patterns (e.g. Fernandez et al. [35]). It will also be advantageous to conduct longitudinal studies to determine what neuromuscular and gait-related factors, including measures of cumulative loading, are most associated with onset and progression of the disease. An important strength of the study was that eligibility for both groups were based on radiographic and symptomatic criteria which minimised the risk of participant misclassification, as it is well known that individuals can have radiological OA and exhibit no symptoms, and vice-versa [36]. 5. Conclusions Despite having a higher body mass, individuals with mild-to-moderate hip OA experience under-loading of the hip joint in the sagittal and transverse planes together with a lesser range of hip joint motion over a longer stance phase, compared to healthy controls. The implications of these altered joint mechanics for the onset and progression of the disease require further investigation.

14

Financial support: Maria Constantinou was supported by a National Health Medical Research Council Postgraduate Research Scholarship, Fellowship Funds Incorporated Ltd Fellowships, Griffith University Postgraduate Scholarship. Conflict of interest: Authors have no conflict of interest. No financial or personal relationships were conducted with individuals or organisations that could inappropriately influence or bias this work. Acknowledgements Funding for this project was provided by a National Health Medical Research Council Postgraduate Scholarship, Australia, a Fellowship Funds Incorporated Ltd Fellowship and a Griffith University Postgraduate scholarship for Maria Constantinou and a Griffith University Strategic Investments Grant for the project. The authors wish to thank Prof. David Lloyd, Prof. Mark Forwood, Prof. Belinda Beck, Dr Gary Shepherd (Qscan Radiology Clinics), Jeremy Higgs, Hoa Hoang, Andrea Miller, Sandra Edwards and the participants for support with the project.

15

References [1] Grazio S. Osteoarthritis-epidemiology, economics and quality of life. Osteoartritisepidemiologija, ekonomski aspekti i kvaliteta zivota. 2005;52:21-9. [2] Lin YC, Davey RC, Cochrane T. Tests for physical function of the elderly with knee and hip osteoarthritis. Scand J Med Sci Sports. 2001;11:280-6. [3] Constantinou M, Barrett R, Brown M, Mills P. Spatial-temporal gait characteristics in individuals with hip osteoarthritis: A systematic literature review and meta-analysis. J Orthop Sports Phys Ther. 2014;44:291-B7. [4] Saxby DJ, Lloyd DG. Osteoarthritis year in review 2016: mechanics. Osteoarthritis Cartilage. 2016. [5] Foucher KC, Schlink BR, Shakoor N, Wimmer MA. Sagittal plane hip motion reversals during walking are associated with disease severity and poorer function in subjects with hip osteoarthritis. J Biomech. 2012;45:1360-5. [6] Shakoor N, Foucher KC, Wimmer MA, Mikolaitis RA, Khandelwal S, Block JA. Gait alterations and their association with radiographic severity and pain in unilateral hip osteoarthritis. Osteoarthritis Cartilage. 2010;18:S67-S8. [7] Kumar D, Wyatt C, Chiba K, Lee S, Nardo L, Link TM, et al. Anatomic correlates of reduced hip extension during walking in individuals with mild-moderate radiographic hip osteoarthritis. J Orthop Res. 2015. [8] Gardiner BS, Woodhouse FG, Besier TF, Grodzinsky AJ, Lloyd DG, Zhang L, et al. Predicting Knee Osteoarthritis. Ann Biomed Eng. 2016;44:222-33. [9] Foucher KC, Hurwitz DE, Wimmer MA. Preoperative gait adaptations persist one year after surgery in clinically well-functioning total hip replacement patients. J Biomech. 2007;40:34327. [10] Hurwitz DE, Hulet CH, Andriacchi TP, Rosenberg AG, Galante JO. Gait compensations in patients with osteoarthritis of the hip and their relationship to pain and passive hip motion. J Orthop Res. 1997;15:629-35.

16

[11] Klausmeier V, Lugade V, Jewett BA, Collis DK, Chou LS. Is there faster recovery with an anterior or anterolateral THA? A pilot study. Clin Orthop Relat Res. 2010;468:533-41. [12] Kiss RM. Effect of walking speed and severity of hip osteoarthritis on gait variability. J Electromyogr Kinesiol. 2010;20:1044-51. [13] Eitzen I, Fernandes L, Nordsletten L, Risberg MA. Sagittal plane gait characteristics in hip osteoarthritis patients with mild to moderate symptoms compared to healthy controls: a cross-sectional study. BMC Musculoskelet Disord. 2012;13:258. [14] Watelain E, Dujardin F, Babier F, Dubois D, Allard P. Pelvic and lower limb compensatory actions of subjects in an early stage of hip osteoarthritis. Arch Phys Med Rehabil. 2001;82:1705-11. [15] Leigh RJ, Osis ST, Ferber R. Kinematic gait patterns and their relationship to pain in mildto-moderate hip osteoarthritis. Clin Biomech. 2016;34:12-7. [16] Hochberg MC. Do risk factors for incident hip osteoarthritis (OA) differ from those for progression of hip OA? [see comment]. J Rheumatol Suppl. 2004;70:6-9. [17] Mahomed NN, Arndt DC, McGrory BJ, Harris WH. The Harris Hip Score: comparison of patient self-report with surgeon assessment. J Arthroplasty. 2001;16:575-80. [18] Auleley GR, Giraudeau B, Dougados M, Ravaud P. Radiographic assessment of hip osteoarthritis progression: impact of reading procedures for longitudinal studies. Ann Rheum Dis. 2000;59:422-7. [19] Altman RD, Bloch DA, Dougados M, Hochberg M, Lohmander S, Pavelka K, et al. Measurement of structural progression in osteoarthritis of the hip: the Barcelona Consensus Group. Osteoarthritis Cartilage. 2004;12:515-24. [20] Lane NE, Nevitt MC, Hochberg MC, Hung YY, Palermo L. Progression of radiographic hip osteoarthritis over eight years in a community sample of elderly white women. Arthritis Rheum. 2004;50:1477-86. [21] Lequesne M, Malghem J, Dion E. The normal hip joint space: variations in width, shape, and architecture on 223 pelvic radiographs. Ann Rheum Dis. 2004;63:1145-51.

17

[22] Altman RD, Gold GE. Atlas of individual radiographic features in osteoarthritis, revised. Osteoarthritis Cartilage. 2007;15 Supplement A:A1-56. [23] Kellgren JH, Lawrence JS. Radiological assessment of osteo-arthrosis. Ann Rheum Dis. 1957;16:494-502. [24] Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods. 2007;39:175-91. [25] Byrd JW, Jones KS. Prospective analysis of hip arthroscopy with 2-year follow-up. Arthroscopy. 2000;16:578-87. [26] Besier TF, Sturnieks DL, Alderson JA, Lloyd DG. Repeatability of gait data using a functional hip joint centre and a mean helical knee axis. J Biomech. 2003;36:1159-68. [27] Harrington ME, Zavatsky AB, Lawson SE, Yuan Z, Theologis TN. Prediction of the hip joint centre in adults, children, and patients with cerebral palsy based on magnetic resonance imaging. J Biomech. 2007;40:595-602. [28] Kainz H, Carty CP, Modenese L, Boyd RN, Lloyd DG. Estimation of the hip joint centre in human motion analysis: a systematic review. Clin Biomech. 2015;30:319-29. [29] de Leva P. Adjustments to Zatsiorsky-Seluyanov's segment inertia parameters. J Biomech. 1996;29:1223-30. [30] Koopman B, Grootenboer HJ, de Jongh HJ. An inverse dynamics model for the analysis, reconstruction and prediction of bipedal walking. J Biomech. 1995;28:1369-76. [31] Barre A, Armand S. Biomechanical ToolKit: Open-source framework to visualize and process biomechanical data. Comput Methods Programs Biomed. 2014;114:80-7. [32] Ornetti P, Laroche D, Morisset C, Beis J, Tavernier C, Maillefert J. Three-dimensional kinematics of the lower limbs in hip osteoarthritis during walking. J Back Musculoskelet Rehabil. 2011;24:201 - 8.

18

[33] Mabuma J, Schwarze M, Hurschler C, Markert B, Ehlers W. Effects of osteoarthritis and pathological walking on contact stresses in femoral cartilage. Biomech Model Mechanobiol. 2015;14:1167-80. [34] Kubota M, Shimada S, Kobayashi S, Sasaki S, Kitade I, Matsumura M, et al. Quantitative gait analysis of patients with bilateral hip osteoarthritis excluding the influence of walking speed. J Orthop Sci. 2007;12:451-7. [35] Fernandez J, Sartori M, Lloyd D, Munro J, Shim V. Bone remodelling in the natural acetabulum is influenced by muscle force-induced bone stress. Int J Numer Meth Biomed. 2014;30:28-41. [36] Kim C, Linsenmeyer KD, Vlad SC, Guermazi A, Clancy MM, Niu J, et al. Prevalence of radiographic and symptomatic hip osteoarthritis in an urban United States community: the Framingham Osteoarthritis Study. Arthritis Rheumatol. 2014;66:3013-7.

19

Figure 1. The ensemble averages for pelvis, hip kinematics during a full stride for individuals in the hip OA and control groups. * Denotes significant difference between groups at p < 0.05. Figure 2. The ensemble averages for internal hip moments during a full stride for individuals in hip OA and control groups. * Denotes significant difference between groups at p < 0.05. † Denotes significant difference between groups after accounting for gait speed (p < 0.05) § Denotes significant difference between groups after accounting for body mass (p < 0.05)

20

Figure 1

21

Figure 2

22

Table 1. Participant Characteristics. Data are mean ± one standard deviation. Hip OA

Control

F, p

n

27 (18Fe, 9M)

26 (18Fe, 8M)

Age (years)

63.2 ± 7.6

59.3 ± 7.6

3.4, 0.071

Height (m)

1.66 ± 0.09

1.69 ± 0.08

0.91, 0.343

Mass (kg)

77.6 ± 14.2

70.5 ± 9.3

4.6, 0.036*

Body Mass Index (kgm-2)

28.0 ± 4.1

24.8 ± 3.0

10.3, 0.002*

Joint Space Width (mm)

2.39 ± 0.98

4.02 ± 0.67

49.6, <0.001*

Modified Harris Hip Score

69.59 ± 12.99

99.85 ± 0.90

140.2, <0.001*

OA: Osteoarthritis, Fe: Females, M: Males, n: Number * Denotes significant difference between groups at p < 0.05

23

Table 2. Spatial-temporal gait characteristics for individuals with hip OA (n = 27) and controls (n = 26). Data are mean ± one standard deviation.

Hip OA

Control

F, p

Gait Speed (self-selected) (m/s)

1.21 ± 0.22

1.39 ± 0.15

12.2, 0.001*

Cadence (steps/minute)

113.7 ± 10.6

117.3 ± 6.8

2.2, 0.145

Step length (m)

0.65 ± 0.09

0.72 ± 0.06

11.0, 0.002*

Step width (m)

0.12 ± 0.04

0.10 ± 0.04

2.0, 0.164

Stride duration (s)

1.06 ± 0.10

1.03 ± 0.06

2.8, 0.100

Stance duration (% stride)

64.4 ± 1.5

62.9 ± 1.2

15.7, <0.001*†

Swing duration (% stride)

35.6 ± 1.5

37.1 ± 1.2

15.3, <0.001*†

Double support duration (% stride)

14.1 ± 1.6

12.9 ± 1.3

9.0, 0.004*

* Denotes significant difference between groups at p < 0.05 † Denotes significant difference between groups after accounting for gait speed at p < 0.05 n: Number

24

Table 3. Peak angles and peak internal joint moments during gait in individuals with hip OA (n = 27) and controls (n = 26). Data are mean ± one standard deviation. Hip OA

Control

F, p

Peak anterior tilt during stance

8.0 ± 5.4

6.8 ± 5.9

0.7, 0.420

Peak superior obliquity during stance

2.7 ± 3.5

5.08 ± 2.7

8.1, 0.006*

Peak flexion during stance

29.1 ± 7.8

29.6 ± 7.3

0.6, 0.813

Peak extension during stance

8.4 ± 7.0

14.2 ± 8.7

7.1, 0.010*

Peak adduction during stance

7.2 ± 4.1

8.3 ± 3.8

1.2, 0.288

Peak extension moment during stance

55.3 ± 24.1

59.7 ± 19.7

0.5, 0.486§

Peak flexion moment during stance

57.2 ± 28.1

71.3 ±28.2

3.1, 0.082§

Peak flexion moment at toe-off

21.7 ± 8.8

30.1 ± 12.6

7.7, 0.008*§

Peak abduction moment at early stance

68.4 ± 17.8

67.5 ± 13.2

0.4, 0.834

Peak abduction moment at mid stance

51.7 ± 16.3

42.5 ± 10.1

5.7, 0.021*

Peak abduction moment at late stance

65.7 ± 18.0

69.1 ± 12.3

0.6, 0.441§

Peak internal rotation moment at foot strike

-0.6 ± 5.1

3.9 ± 6.6

7.8, 0.008*†§

Pelvis Angle (degrees)

Hip Angle (degrees)

Internal Hip Moment (Nm)

* Denotes significant difference between groups at p < 0.05 † Denotes significant difference between groups after accounting for gait speed (p < 0.05) § Denotes significant difference between groups after accounting for body mass (p < 0.05) n: Number

25