Improvements in Knee Biomechanics During Walking Are Associated With Increased Physical Activity After Total Knee Arthroplasty John B. Arnold,1 Shylie Mackintosh,2 Timothy S. Olds,1 Sara Jones,2 Dominic Thewlis1,3 1 Alliance for Research in Exercise, Nutrition and Activity (ARENA), School of Health Sciences, Sansom Institute for Health Research, University of South Australia, Adelaide, Australia, 2International Centre for Allied Health Evidence (iCAHE), School of Health Sciences, Sansom Institute for Health Research, University of South Australia, Adelaide, Australia, 3Centre for Orthopaedic and Trauma Research, University of Adelaide, Adelaide, Australia
Received 6 February 2015; accepted 16 June 2015 Published online 14 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22969
ABSTRACT: Total knee arthroplasty (TKA) in people with knee osteoarthritis increases knee-specific and general physical function, but it has not been established if there is a relationship between changes in these elements of functional ability. This study investigated changes and relationships between knee biomechanics during walking, physical activity, and use of time after TKA. Fifteen people awaiting TKA underwent 3D gait analysis before and six months after surgery. Physical activity and use of time were determined in free-living conditions from a high resolution 24-h activity recall. After surgery, participants displayed significant improvements in sagittal plane knee biomechanics and improved their physical activity profiles, standing for 105 more minutes (p ¼ 0.001) and performing 64 min more inside chores on average per day (p ¼ 0.008). Changes in sagittal plane knee range of motion (ROM) and peak knee flexion positively correlated with changes in total daily energy expenditure, time spent undertaking moderate to vigorous physical activity, inside chores and passive transport (r ¼ 0.52–0.66, p ¼ 0.005–0.047). Restoration of knee function occurs in parallel and is associated with improvements in physical activity and use of time after TKA. Increased functional knee ROM is required to support improvements in total and context specific physical activity. ß 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 33:1818–1825, 2015. Keywords: knee; gait analysis; biomechanics; physical activity; knee replacement
Total knee arthroplasty (TKA) is an effective procedure for improving symptoms of end-stage knee osteoarthritis (OA) that are unresponsive to conservative treatment.1 The number of TKA procedures in developed countries has been increasing over the past 10 years,2–4 and is projected to continue to rise because of the increasing prevalence of symptomatic knee OA, largely related to population ageing and rising obesity levels.5 Total knee arthroplasty aims to improve knee function by increased range of motion and reduction of abnormal joint mechanics. Assessment of knee range of motion is commonly performed post-operatively and improving the range of knee flexion or extension is often a target of rehabilitation before and after discharge from hospital.6,7 Improvement in knee motion after TKA is considered important as it is assumed to relate to increased functional ability after surgery, particularly for common activities of daily living that require knee flexion such as walking, climbing stairs, kneeling, or squatting.8,9 Numerous methods have been used to quantify changes in knee biomechanics and physical activity after TKA in laboratory and free living conditions. These include three dimensional (3D) gait analysis,10 accelerometry11–13, and self-report questionnaires.1 While studies have separately tracked changes in knee biomechanics and physical activity after TKA, few have investigated the relationship between these two elements of functional ability. Of these studies, track-
ing of physical activity has been limited to simple selfreport questionnaires,14–18 which lack the ability to determine the type and frequency of activities undertaken. Information about the context of the physical activity using high-resolution activity recall may improve the understanding of what type and how much activity may be related to changes in knee function after TKA. Instruments such as the Multimedia Activity Recall for Children and Adults (MARCA) can create detailed activity profiles for individuals over extended periods, and display good convergent validity with accelerometry.19,20 The synthesis of high fidelity biomechanical data with information on physical activity in free-living conditions provides a new approach to elucidating how function, mobility and activity are related over time after surgical intervention. Understanding what factors are related to the change across this transition has the potential to improve the management of patients undergoing TKA surgery and inform the development of interventions aimed to increase postoperative physical activity. Therefore, the primary aim of this study was to investigate the changes and relationships between knee biomechanics during walking, physical activity and use of time in free-living conditions after TKA. The secondary aim was to investigate if there is a relationship between changes in knee symptoms and postoperative changes in physical activity.
PATIENTS AND METHODS Conflict of Interest: None to declare. Correspondence to: John Arnold (T: þ 61 8 8302 1207; F: þ 61 8 8302 2766; E-mail: [email protected]) # 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
1818
JOURNAL OF ORTHOPAEDIC RESEARCH DECEMBER 2015
Participants Participants were enrolled prospectively in this single group cohort study (Level 3) and were on the waiting list for primary unilateral TKA at the Department of Orthopaedics at The
BIOMECHANICS AND ACTIVITY AFTER TKA
Queen Elizabeth Hospital in Adelaide, Australia. All were required to be over 50 years of age and met the American College of Rheumatology classification criteria for clinical and radiographic knee OA.21 Grading of OA severity was undertaken according to the Kellgren-Lawrence (KL) system.22 Exclusion criteria were: a history of other lower limb joint replacement; major orthopaedic surgery of the back or lower limbs (including high tibial osteotomy); corticosteroid injection to the knee within the past six months; systemic arthritic condition; a cardiac, neurological or musculoskeletal condition affecting gait; inability to walk household distances; and cognitive disorder or inability to understand English. The contralateral knee joint was not under consideration for joint replacement surgery in any participant. The affected limb in this study was defined as the knee scheduled for surgery. This study was approved by The Queen Elizabeth Hospital and University of South Australia Human Research Ethics Committees. All participants gave written informed consent prior to their involvement. Surgical Intervention and Rehabilitation All participants underwent a midline incision with a medial parapatellar approach to surgery by one of four consultant surgeons. No intra-operative complications were reported. Five different prosthesis designs were used; Genesis IITM 1 (Smith & Nephew Inc, Memphis, TN; n ¼ 7), the Vanguard (Biomet Inc, Warsaw, IN; n ¼ 5), the Triathlon (Stryker 1 Orthopaedics, Mahwah, NJ; n ¼ 1), the Nexgen Legacy 1 TM (Zimmer Inc, Warsaw, IN; n ¼ 1) and the LCS Complete (DePuy Orthopaedics Inc, Warsaw, IN; n ¼ 1). All tibial components were cemented, with nine patients having uncemented femoral components. In-patient physiotherapy consisted of weight-bearing or partial weight bearing as tolerated beginning the day after surgery, progressing from assisted to unassisted standing and walking. Non-weight bearing lower limb exercises included inner range quadriceps strengthening (active movement of knee from approximately 20˚ flexion to full extension), straight leg raise and active knee flexion/extension exercises to increase ROM as deemed appropriate by the physiotherapist. The median time spent in hospital was 6 days (range 4–11 days) and after discharge patients continued with weight bearing as tolerated. Gait Analysis Gait analysis was performed on all participants preoperatively and six months after TKA. Three-dimensional kinematics and ground reaction forces were recorded during walking using 12 VICON MX-F20 cameras (Vicon Metrics, Oxford, UK) and two floor-embedded Kistler force platforms (9281B, Kistler Instrument Corporation, Switzerland) at 100 Hz and 400 Hz, respectively. Walking speed was measured by two infrared photocells (Speed Light V2, Swift Performance Equipment, Queensland, Australia). A lower limb marker set was used where retro-reflective surface markers were placed bilaterally on the anterior superior iliac spine, posterior superior iliac spine, greater trochanter (not used for tracking), medial and lateral femoral epicondyle, medial and lateral malleolus, posterior aspect of the calcaneus, first and fifth metatarsal heads and dorsal foot (tracking marker). Rigid clusters of four markers were affixed to the thigh and shank segments. A static trial with each participant standing in relaxed bipedal stance with both feet aligned with the long axis of the laboratory coordinate system was captured to define the position and orientation in space of each body
1819
segment and to define the joint centres.23 Each participant completed three walking trials along a 15 m walkway at a self-selected speed. Kinematic and kinetic data were processed in Visual3D (v4.0, C-Motion) and filtered with a fourth-order low-pass Butterworth filter with cut-off frequencies of 6 and 25 Hz, respectively.24 A kinematic model of the lower limbs was constructed with included a pelvis, thigh, shank and foot segments. The hip joint centre was defined based on the regression model of Bell et al.,25 the knee joint centre as the midpoint of the medial and lateral femoral epicondyles and the ankle joint from the midpoint of the malleoli markers. Joint angles were computed using the joint coordinate system approach.26 External joint moments were computed using inverse dynamics (resolved in distal segment coordinate system) and were normalised to body weight and height (%BW Ht) and reported as the mean of the three trials per participant. Data were time normalised to 0 to 100% of the gait cycle. The variables of interest in this study were walking speed, the peak knee flexion moment during loading response (initial contact to first peak knee flexion) as a measure of loading confidence, knee flexion excursion from initial contact to peak flexion, total sagittal plane knee ROM during the gait cycle, and peak knee flexion during swing phase. Physical Activity and Use of Time Recall Use of time data were collected using a computerised 24-h activity diary; the Multimedia Activity Recall for Children and Adults (MARCA).19,20 This tool asks participants to recall all activities undertaken in the previous 24 hours in time-slices of five minutes or more. Participants can choose from over 500 different activities (e.g., watching television, brushing teeth, driving a car). Each activity in the MARCA is assigned a metabolic equivalent of tasks value (MET) based on the Compendium of Physical Activities so energy expenditure can be estimated.27,28 The time spent (in minutes) performing specific activities can also be calculated, including minutes spent within “activity sets”, such as inside chores, self-care and transportation. While designed as a self-report instrument, the MARCA can also be administered by face-to-face or telephone interview. By combining a 24-h activity recall in a time-diary format, a compendium of energy costs and an analytical module, the MARCA draws on the strengths of both use-of-time and physical activity methodologies, resulting in a high-resolution instrument that is capable of both use-of-time and energy expenditure analysis. Evaluation of the MARCA in adults has shown high test-retest reliability (ICC ¼ 0.992– 0.997) and moderate to strong validity based on correlations between physical activity level (PAL) and accelerometer counts per minute (rho ¼ 0.72)20 as well as total daily energy expenditure (TDEE) and doubly-labelled water (rho ¼ 0.70).29 Preoperatively and six months postoperatively the MARCA was administered over the telephone by a trained interviewer. Participants were asked to recall a total of four days of activity which included at least one weekday and one weekend day. Preoperatively, data were collected within eight weeks of surgery, but not within the week prior to surgery as participants were likely to have altered their use of time in preparation for hospital admission. Postoperative data collection was scheduled for within two weeks of the six month time-point. JOURNAL OF ORTHOPAEDIC RESEARCH DECEMBER 2015
1820
ARNOLD ET AL.
The following MARCA variables were calculated (Table 1): mean minutes spent at 1-1.9 METs (very light/ sedentary physical activity (VLPA); mean minutes spent at 3 METs (moderate to vigorous physical activity (MVPA); mean total daily energy expenditure (TDEE), calculated by multiplying the duration of each activity with the associated energy cost (MET.min); and mean time spent standing and sitting per day (minutes). The time spent performing specific activities, relevant to middle- and older-aged adults, was also extracted. These included indoor household chores and time spent in passive transport (riding or driving motorised transport).30 The mean value reported for each variable at each time-point represents the average of the four-day recall period, with data weighted 5:2 (weekday: weekend) to capture typical weekly patterns. An overview of the workflow for data processing in this study is illustrated in Figure 1. Knee Symptoms The Western Ontario & McMaster Universities Osteoarthritis Index (WOMAC) (five point Likert-type format) was completed by participants with knee OA.31 The WOMAC (0– 96 points) includes three subscales: pain (0–20 points), stiffness (0–8 points) and activities of daily living (0-68 points) with lower scores indicating less knee pain, stiffness and functional limitation. This questionnaire has been widely used in adults undergoing TKA,1 has good internal consistency and established reliability.32,33 Statistical Analysis Prior to undertaking inferential statistical testing, distribution of data sets was assessed for normality using ShapiroWilks tests. All data were normally distributed; therefore relationships between changes in sagittal plane knee kinematics and external joint moments, walking speed, physical activity and use of time data from the MARCA and the WOMAC subscales were evaluated with Pearson’s correlation coefficients (r). Changes in knee biomechanics, walking patterns, physical activity and use of time between the preoperative and six months postoperative time-points were evaluated with paired t-tests. P-values less than 0.05 were considered statistically significant.
RESULTS Participants From the 61 patients on the surgical waiting list for primary TKA between August 2012 and May 2013, 21
did not meet the study inclusion criteria, 18 declined to participate and two were taken off the waiting list. Twenty participants completed the preoperative gait analysis on average 9 weeks before surgery (range 2–16 weeks). One participant withdrew from the study as they did not proceed to surgery and two participants withdrew after surgery because of unrelated medical complications. Seventeen participants completed the six month gait analysis (range (4 to þ5 weeks from time-point). Of these, 15 participants (mean age 67.8 years SD 10.4, height 1.64 m SD 0.1, body mass 85.4 kg SD 15.1, BMI 31.8 kg/m2 SD 5.5, 6 M: 9F, KL grade 3: 4, KL grade 4: 11) had full physical activity and use-of-time data at both timepoints (two participants had partial data as they were not available for all sessions), and were included in the analysis. As expected, six months after surgery there was a significant mean improvement in pain (7.9 points, SD 3.9, range 0 to 14, p < 0.001), stiffness (3.4 points, SD 2.1, range 5 to 47, p < 0.001) and activity limitation (26.1 points SD 13.1, range 5 to 47, p < 0.001) subscales and total WOMAC score (37.5 points SD 17.8, range 10 to 67, p < 0.001). 66. Knee Biomechanics and Spatiotemporal Gait Patterns At six months after TKA, participants had significant increases in peak knee flexion (29.3˚, range 18.7˚–49.4˚ p ¼
Received 6 February 2015; accepted 16 June 2015 Published online 14 July 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22969
ABSTRACT: Total knee arthroplasty (TKA) in people with knee osteoarthritis increases knee-specific and general physical function, but it has not been established if there is a relationship between changes in these elements of functional ability. This study investigated changes and relationships between knee biomechanics during walking, physical activity, and use of time after TKA. Fifteen people awaiting TKA underwent 3D gait analysis before and six months after surgery. Physical activity and use of time were determined in free-living conditions from a high resolution 24-h activity recall. After surgery, participants displayed significant improvements in sagittal plane knee biomechanics and improved their physical activity profiles, standing for 105 more minutes (p ¼ 0.001) and performing 64 min more inside chores on average per day (p ¼ 0.008). Changes in sagittal plane knee range of motion (ROM) and peak knee flexion positively correlated with changes in total daily energy expenditure, time spent undertaking moderate to vigorous physical activity, inside chores and passive transport (r ¼ 0.52–0.66, p ¼ 0.005–0.047). Restoration of knee function occurs in parallel and is associated with improvements in physical activity and use of time after TKA. Increased functional knee ROM is required to support improvements in total and context specific physical activity. ß 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 33:1818–1825, 2015. Keywords: knee; gait analysis; biomechanics; physical activity; knee replacement
Total knee arthroplasty (TKA) is an effective procedure for improving symptoms of end-stage knee osteoarthritis (OA) that are unresponsive to conservative treatment.1 The number of TKA procedures in developed countries has been increasing over the past 10 years,2–4 and is projected to continue to rise because of the increasing prevalence of symptomatic knee OA, largely related to population ageing and rising obesity levels.5 Total knee arthroplasty aims to improve knee function by increased range of motion and reduction of abnormal joint mechanics. Assessment of knee range of motion is commonly performed post-operatively and improving the range of knee flexion or extension is often a target of rehabilitation before and after discharge from hospital.6,7 Improvement in knee motion after TKA is considered important as it is assumed to relate to increased functional ability after surgery, particularly for common activities of daily living that require knee flexion such as walking, climbing stairs, kneeling, or squatting.8,9 Numerous methods have been used to quantify changes in knee biomechanics and physical activity after TKA in laboratory and free living conditions. These include three dimensional (3D) gait analysis,10 accelerometry11–13, and self-report questionnaires.1 While studies have separately tracked changes in knee biomechanics and physical activity after TKA, few have investigated the relationship between these two elements of functional ability. Of these studies, track-
ing of physical activity has been limited to simple selfreport questionnaires,14–18 which lack the ability to determine the type and frequency of activities undertaken. Information about the context of the physical activity using high-resolution activity recall may improve the understanding of what type and how much activity may be related to changes in knee function after TKA. Instruments such as the Multimedia Activity Recall for Children and Adults (MARCA) can create detailed activity profiles for individuals over extended periods, and display good convergent validity with accelerometry.19,20 The synthesis of high fidelity biomechanical data with information on physical activity in free-living conditions provides a new approach to elucidating how function, mobility and activity are related over time after surgical intervention. Understanding what factors are related to the change across this transition has the potential to improve the management of patients undergoing TKA surgery and inform the development of interventions aimed to increase postoperative physical activity. Therefore, the primary aim of this study was to investigate the changes and relationships between knee biomechanics during walking, physical activity and use of time in free-living conditions after TKA. The secondary aim was to investigate if there is a relationship between changes in knee symptoms and postoperative changes in physical activity.
PATIENTS AND METHODS Conflict of Interest: None to declare. Correspondence to: John Arnold (T: þ 61 8 8302 1207; F: þ 61 8 8302 2766; E-mail: [email protected]) # 2015 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
1818
JOURNAL OF ORTHOPAEDIC RESEARCH DECEMBER 2015
Participants Participants were enrolled prospectively in this single group cohort study (Level 3) and were on the waiting list for primary unilateral TKA at the Department of Orthopaedics at The
BIOMECHANICS AND ACTIVITY AFTER TKA
Queen Elizabeth Hospital in Adelaide, Australia. All were required to be over 50 years of age and met the American College of Rheumatology classification criteria for clinical and radiographic knee OA.21 Grading of OA severity was undertaken according to the Kellgren-Lawrence (KL) system.22 Exclusion criteria were: a history of other lower limb joint replacement; major orthopaedic surgery of the back or lower limbs (including high tibial osteotomy); corticosteroid injection to the knee within the past six months; systemic arthritic condition; a cardiac, neurological or musculoskeletal condition affecting gait; inability to walk household distances; and cognitive disorder or inability to understand English. The contralateral knee joint was not under consideration for joint replacement surgery in any participant. The affected limb in this study was defined as the knee scheduled for surgery. This study was approved by The Queen Elizabeth Hospital and University of South Australia Human Research Ethics Committees. All participants gave written informed consent prior to their involvement. Surgical Intervention and Rehabilitation All participants underwent a midline incision with a medial parapatellar approach to surgery by one of four consultant surgeons. No intra-operative complications were reported. Five different prosthesis designs were used; Genesis IITM 1 (Smith & Nephew Inc, Memphis, TN; n ¼ 7), the Vanguard (Biomet Inc, Warsaw, IN; n ¼ 5), the Triathlon (Stryker 1 Orthopaedics, Mahwah, NJ; n ¼ 1), the Nexgen Legacy 1 TM (Zimmer Inc, Warsaw, IN; n ¼ 1) and the LCS Complete (DePuy Orthopaedics Inc, Warsaw, IN; n ¼ 1). All tibial components were cemented, with nine patients having uncemented femoral components. In-patient physiotherapy consisted of weight-bearing or partial weight bearing as tolerated beginning the day after surgery, progressing from assisted to unassisted standing and walking. Non-weight bearing lower limb exercises included inner range quadriceps strengthening (active movement of knee from approximately 20˚ flexion to full extension), straight leg raise and active knee flexion/extension exercises to increase ROM as deemed appropriate by the physiotherapist. The median time spent in hospital was 6 days (range 4–11 days) and after discharge patients continued with weight bearing as tolerated. Gait Analysis Gait analysis was performed on all participants preoperatively and six months after TKA. Three-dimensional kinematics and ground reaction forces were recorded during walking using 12 VICON MX-F20 cameras (Vicon Metrics, Oxford, UK) and two floor-embedded Kistler force platforms (9281B, Kistler Instrument Corporation, Switzerland) at 100 Hz and 400 Hz, respectively. Walking speed was measured by two infrared photocells (Speed Light V2, Swift Performance Equipment, Queensland, Australia). A lower limb marker set was used where retro-reflective surface markers were placed bilaterally on the anterior superior iliac spine, posterior superior iliac spine, greater trochanter (not used for tracking), medial and lateral femoral epicondyle, medial and lateral malleolus, posterior aspect of the calcaneus, first and fifth metatarsal heads and dorsal foot (tracking marker). Rigid clusters of four markers were affixed to the thigh and shank segments. A static trial with each participant standing in relaxed bipedal stance with both feet aligned with the long axis of the laboratory coordinate system was captured to define the position and orientation in space of each body
1819
segment and to define the joint centres.23 Each participant completed three walking trials along a 15 m walkway at a self-selected speed. Kinematic and kinetic data were processed in Visual3D (v4.0, C-Motion) and filtered with a fourth-order low-pass Butterworth filter with cut-off frequencies of 6 and 25 Hz, respectively.24 A kinematic model of the lower limbs was constructed with included a pelvis, thigh, shank and foot segments. The hip joint centre was defined based on the regression model of Bell et al.,25 the knee joint centre as the midpoint of the medial and lateral femoral epicondyles and the ankle joint from the midpoint of the malleoli markers. Joint angles were computed using the joint coordinate system approach.26 External joint moments were computed using inverse dynamics (resolved in distal segment coordinate system) and were normalised to body weight and height (%BW Ht) and reported as the mean of the three trials per participant. Data were time normalised to 0 to 100% of the gait cycle. The variables of interest in this study were walking speed, the peak knee flexion moment during loading response (initial contact to first peak knee flexion) as a measure of loading confidence, knee flexion excursion from initial contact to peak flexion, total sagittal plane knee ROM during the gait cycle, and peak knee flexion during swing phase. Physical Activity and Use of Time Recall Use of time data were collected using a computerised 24-h activity diary; the Multimedia Activity Recall for Children and Adults (MARCA).19,20 This tool asks participants to recall all activities undertaken in the previous 24 hours in time-slices of five minutes or more. Participants can choose from over 500 different activities (e.g., watching television, brushing teeth, driving a car). Each activity in the MARCA is assigned a metabolic equivalent of tasks value (MET) based on the Compendium of Physical Activities so energy expenditure can be estimated.27,28 The time spent (in minutes) performing specific activities can also be calculated, including minutes spent within “activity sets”, such as inside chores, self-care and transportation. While designed as a self-report instrument, the MARCA can also be administered by face-to-face or telephone interview. By combining a 24-h activity recall in a time-diary format, a compendium of energy costs and an analytical module, the MARCA draws on the strengths of both use-of-time and physical activity methodologies, resulting in a high-resolution instrument that is capable of both use-of-time and energy expenditure analysis. Evaluation of the MARCA in adults has shown high test-retest reliability (ICC ¼ 0.992– 0.997) and moderate to strong validity based on correlations between physical activity level (PAL) and accelerometer counts per minute (rho ¼ 0.72)20 as well as total daily energy expenditure (TDEE) and doubly-labelled water (rho ¼ 0.70).29 Preoperatively and six months postoperatively the MARCA was administered over the telephone by a trained interviewer. Participants were asked to recall a total of four days of activity which included at least one weekday and one weekend day. Preoperatively, data were collected within eight weeks of surgery, but not within the week prior to surgery as participants were likely to have altered their use of time in preparation for hospital admission. Postoperative data collection was scheduled for within two weeks of the six month time-point. JOURNAL OF ORTHOPAEDIC RESEARCH DECEMBER 2015
1820
ARNOLD ET AL.
The following MARCA variables were calculated (Table 1): mean minutes spent at 1-1.9 METs (very light/ sedentary physical activity (VLPA); mean minutes spent at 3 METs (moderate to vigorous physical activity (MVPA); mean total daily energy expenditure (TDEE), calculated by multiplying the duration of each activity with the associated energy cost (MET.min); and mean time spent standing and sitting per day (minutes). The time spent performing specific activities, relevant to middle- and older-aged adults, was also extracted. These included indoor household chores and time spent in passive transport (riding or driving motorised transport).30 The mean value reported for each variable at each time-point represents the average of the four-day recall period, with data weighted 5:2 (weekday: weekend) to capture typical weekly patterns. An overview of the workflow for data processing in this study is illustrated in Figure 1. Knee Symptoms The Western Ontario & McMaster Universities Osteoarthritis Index (WOMAC) (five point Likert-type format) was completed by participants with knee OA.31 The WOMAC (0– 96 points) includes three subscales: pain (0–20 points), stiffness (0–8 points) and activities of daily living (0-68 points) with lower scores indicating less knee pain, stiffness and functional limitation. This questionnaire has been widely used in adults undergoing TKA,1 has good internal consistency and established reliability.32,33 Statistical Analysis Prior to undertaking inferential statistical testing, distribution of data sets was assessed for normality using ShapiroWilks tests. All data were normally distributed; therefore relationships between changes in sagittal plane knee kinematics and external joint moments, walking speed, physical activity and use of time data from the MARCA and the WOMAC subscales were evaluated with Pearson’s correlation coefficients (r). Changes in knee biomechanics, walking patterns, physical activity and use of time between the preoperative and six months postoperative time-points were evaluated with paired t-tests. P-values less than 0.05 were considered statistically significant.
RESULTS Participants From the 61 patients on the surgical waiting list for primary TKA between August 2012 and May 2013, 21
did not meet the study inclusion criteria, 18 declined to participate and two were taken off the waiting list. Twenty participants completed the preoperative gait analysis on average 9 weeks before surgery (range 2–16 weeks). One participant withdrew from the study as they did not proceed to surgery and two participants withdrew after surgery because of unrelated medical complications. Seventeen participants completed the six month gait analysis (range (4 to þ5 weeks from time-point). Of these, 15 participants (mean age 67.8 years SD 10.4, height 1.64 m SD 0.1, body mass 85.4 kg SD 15.1, BMI 31.8 kg/m2 SD 5.5, 6 M: 9F, KL grade 3: 4, KL grade 4: 11) had full physical activity and use-of-time data at both timepoints (two participants had partial data as they were not available for all sessions), and were included in the analysis. As expected, six months after surgery there was a significant mean improvement in pain (7.9 points, SD 3.9, range 0 to 14, p < 0.001), stiffness (3.4 points, SD 2.1, range 5 to 47, p < 0.001) and activity limitation (26.1 points SD 13.1, range 5 to 47, p < 0.001) subscales and total WOMAC score (37.5 points SD 17.8, range 10 to 67, p < 0.001). 66. Knee Biomechanics and Spatiotemporal Gait Patterns At six months after TKA, participants had significant increases in peak knee flexion (29.3˚, range 18.7˚–49.4˚ p ¼
