- Email: [email protected]
Changes in hip mechanics during gait modification to reduce knee abduction moment
Journal of Biomechanics xxx (xxxx) xxx
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Changes in hip mechanics during gait modification to reduce knee abduction moment Sarah Kettlety a, Bryndan Lindsey a, Oladipo Eddo a, Matt Prebble a, Shane Caswell a, Nelson Cortes a,b,⇑ a b
Sports Medicine Assessment, Research & Testing (SMART) Laboratory, George Mason University, School of Kinesiology, Manassas, VA, United States Department of Bioengineering, George Mason University, Fairfax, VA, United States
a r t i c l e
i n f o
Article history: Accepted 12 November 2019 Available online xxxx Keywords: Knee osteoarthritis Gait retraining Real-time biofeedback Hip abduction moment
a b s t r a c t First peak knee abduction moment (KAM) has been associated with the severity and progression of knee osteoarthritis (KOA). Gait modifications, including lateral trunk lean (TL), medial knee thrust (MKT), and reduced foot progression (FP) have decreased KAM. However, their effects on the hip joint are poorly understood. Reduced hip abduction moment has been found to be predictive of KOA progression and has been hypothesized to represent a decreased demand on the hip musculature. Lack of studies investigating changes in hip mechanics as a result of gait modification limits our understanding of their cumulative benefit, therefore, we investigated the effects of TL, MKT, and FP on internal hip abduction moment as well as rate change in net joint reaction force. Using real-time visual biofeedback, five trials were completed for each modification. Each modification target range was individualized to 3–5 SD greater (TL and FP) or lesser (MKT) than the participants mean baseline value. Kinematics and kinetics at the hip and knee were calculated at first peak KAM. Trunk lean and MKT decreased hip abduction moment compared to baseline (p < 0.001). Trunk lean increased rate change in net joint reaction force at both the hip (p < 0.001) and knee (p < 0.001) compared to baseline. Additional research is needed to fully understand the effect of gait modifications in a clinical population, particularly the relationship between hip abduction moments and KOA progression. Although interventions such as MKT and TL can be successful in reducing KAM, their effects on hip abduction moment should be considered before clinical implementation. Ó 2019 Elsevier Ltd. All rights reserved.
1. Introduction The severity (Sharma et al., 1998), symptoms (Amin et al., 2004), and progression (Miyazaki et al., 2002) of knee osteoarthritis (KOA) have been related to increased frontal plane knee moment. The first peak internal knee abduction moment (KAM) is a commonly reported surrogate measure of joint loading (Eddo et al., 2017; Richards et al., 2017). Gait modifications utilizing real-time biofeedback (RTB) have been effective in reducing KAM in both healthy and KOA populations (Eddo et al., 2017; Richards et al., 2017). For example, medial knee thrust (MKT), lateral trunk lean (TL), and reduced foot progression (FP) have shown reductions in KAM of up to 38% (Ferrigno et al., 2016), 65% (Mündermann et al., 2008), and 20% (Shull et al., 2013b) respectively.
⇑ Corresponding author at: Sports Medicine Assessment, Research & Testing (SMART) Laboratory, George Mason University, 10890 George Mason Circle, KJH 201E, MSN 4E5 Manassas, VA 20110, United States. E-mail address: [email protected] (N. Cortes).
Although these modifications have shown themselves able to reduce KAM, their effects on other lower extremity joints remain unclear (Simic et al., 2011). Patients with KOA have lower hip abduction moments compared to healthy counterparts (Astephen et al., 2008; Briem & Snyder-Mackler, 2009) with more severe KOA corresponding to lower hip abduction moments (Astephen et al., 2008; Mündermann et al., 2005). These results suggest that other mechanical measures, such as hip abduction moment, are also important in addition to KAM. The goal of gait modification is to decrease pain, improve function, and slow the progression of KOA. This can be achieved by various mechanisms including shifting the center of mass (COM) towards the symptomatic knee with TL (Hunt et al., 2010; Mündermann et al., 2008; Simic et al., 2011), or shifting of the center of pressure laterally (Shull et al., 2013a) with concomitant medialization of the symptomatic knee joint (Reeves and Bowling, 2011) seen during FP and MKT. Both mechanisms have the effect of lateralizing the GRF vector, reducing frontal plane moment arm and subsequent knee abduction moment. Lateralization of the GRF vector not only affects the knee joint, as limited
https://doi.org/10.1016/j.jbiomech.2019.109509 0021-9290/Ó 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
2
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
evidence has demonstrated an association between decreased KAM and reduced hip abduction moment (Astephen et al., 2008; Mündermann et al., 2005). Lack of studies investigating changes in hip mechanics with gait modification limits our understanding of their cumulative benefit. The available literature suggests that TL significantly decreases hip abduction moment (Briem & Snyder-Mackler, 2009; Mündermann et al., 2008), and increases the rate change of net joint reaction forces (Mündermann et al., 2005). The effects of other commonly studied gait modifications remain unknown. Therefore, the purpose of this study was to investigate how FP, MKT, and TL gait modifications affect internal hip abduction moment. We hypothesized that all gait modifications would significantly reduce KAM and hip abduction moment from baseline.
mean and standard deviation (SD) for three modification parameters: frontal plane trunk and knee angle, and transverse plane foot angle. Trunk angle was defined as the frontal plane deviation of the trunk segment represented by the right scapula, 10th thoracic, and left/right lower back markers from the vertical laboratory axis (Hunt et al., 2011). Knee angle was defined as the frontal plane knee angle created between the thigh and shank segments (Barrios et al., 2010). Foot angle was found as the offset between the lines formed by the posterior calcaneus and 2nd metatarsophalangeal joint markers, and the anterior-posterior laboratory axis (Shull et al., 2013a). Increased trunk lean to the dominant limb was quantified as positive, increased knee abduction as negative, and reduced foot progression angle as positive. 2.4. Gait modification trials
2. Methods 2.1. Participants Twenty healthy participants (26.7 ± 4.7 years, 1.75 ± 0.1 m, 73.4 ± 12.4 kg) volunteered for this study after giving informed consent approved by the Institution Review Board. A within-group single-session repeated measures design was used to compare joint kinematics and kinetics of participants’ dominant limb across gait conditions. Dominant limb was defined as the preferred leg in a kicking task (Cortes et al., 2012). Eighteen participants were right leg dominant, two left. Participants were only eligible if they were free from any knee, hip, or back pain that required treatment within the prior 6 months and no history of lower limb or back surgery. Exclusion criteria included any neurological or musculoskeletal impairment that would affect gait or any cognitive impairment that would inhibit motor learning. 2.2. Instrumentation Fifty three retroreflective markers were attached to the trunk and lower extremities of participants (Fig. 1). Eight high-speed motion analysis cameras (Vicon, Oxford, England) sampling at 200 Hz tracked marker trajectory. Ground reaction force was acquired using four floor embedded force plates aligned in a 2.4 m long row sampling at 1000 Hz (Bertec, Columbus, OH). A static calibration trial was collected by having participants stand on a force plate with both feet parallel to the anterior-posterior axis of the laboratory. Participants also performed a dynamic calibration to estimate hip joint center by completing three clockwise rotations of the pelvis (Schwartz and Rozumalski, 2005). Calibration makers were removed for walking trials. From the static trial, a kinematic model was created for each participant using Visual 3D software (C-Motion, Germantown MD, USA) which included the trunk, pelvis, and bilateral thigh, shank, and foot segments. 2.3. Baseline trials For all trials, participants walked along a 6-meter long walkway aligned parallel with the anterior – posterior axis of the laboratory at a self-selected speed. Tape was placed at the lateral borders of the walkway in line with the edges of the force plates. Participants were instructed to stay within the tape during trials to prevent deviation from a straight walking path. At the end of each trial, participants returned to the beginning of the walkway, starting at the same point as the prior trial. Timing gates (Brower Timing Systems, Draper UT, USA) positioned at each end of the force plates recorded walking speed during 10 baseline trials. For a trial to be successful, one full contact on a force plate with the dominant foot was required. Recorded data were exported to Visual 3D to calculate
Five trials were completed for each modification using visual RTB, which was delivered in the form of a line graph projected on the wall in front of the walkway (Fig. 2). The graph displayed the angle of the current modification parameter over stance and was updated after each step of the dominant limb. A bandwidth representing a range of 3–5 SD greater (FP and TL) or lesser (MKT) than baseline was displayed on the graph. Participants were instructed to walk so that the line representing the gait parameter fell within the bandwidth. Modification trials were completed in the following order: TL, MKT, and FP. Participants were provided standardized verbal instructions before implementing each modification and completed as many practice trials as needed to familiarize themselves with each modification. Additional verbal feedback was provided during practice trials as needed. Successful trials required one full foot contact of the dominant limb within a force plate, average gait speed of ±5% relative to baseline, and that the line representing the target modification parameter fell within the highlighted bandwidth between approximately 20% to 40% of stance. This last measure of trial success was chosen as some participants found it difficult to stay within the bandwidth during the entire stance period during pilot trials. As we were interested in reducing the first peak of the adduction moment, we counted trials as successful if participants could modify their kinematics at approximately 30% of stance, where peak KAM typically occurs (Hunt et al., 2008; Simic et al., 2011). Only successful trials counted towards the 5 required for each modification. 2.5. Data processing The kinematic model created in Visual 3D was used to quantify the motion at the hip, knee, and ankle joints with rotations being expressed relative to the static trial. A cardan angle sequence was used to calculate joint angles (Grood and Suntay, 1983) and standard inverse dynamics analysis was conducted to synthesize the trajectory and vertical ground reaction force data for internal joint moment estimation at the knee and hip. Although external joint moments are most commonly reported in KOA literature, internal moments resist the action of external moments and can be thought of as equal but opposite in sign (Butler et al., 2009). Joint kinematics and kinetics were smoothed using a low-pass Butterworth filter with a cut off frequency of 8 Hz to reduce the effects of artifacts based on results from residual analysis (Winter, 2009). Joint angles were measured in degrees, and internal joint moments were normalized to mass and height (Nm/Kgm). Ground reaction force data were normalized to body weight and gait trials were normalized to 100 percent of stance. Modification parameters were calculated as the average across stance. First peak for knee and hip abduction moment and hip flexion moment were defined as the peak minimum and maximum
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
3
Fig. 1. Experimental marker placement. Four tracking clusters (18 markers) were placed on the lateral aspect of each thigh and shank. Additional tracking markers (22 markers) were attached to the manubrium, 7th cervical vertebrae, right scapula, 10th thoracic vertebrae, and bilaterally to the following locations: posterior and lateral calcaneus, 5th distal metatarsal, 1st proximal metatarsal, 2nd metatarsophalangeal joint, tibial tuberosity, lateral iliac spine, posterior superior iliac spine, and acromion. Three tracking markers, arranged to form a triangular cluster, were attached to the lower back. Ten additional calibration markers were attached bilaterally to the following anatomical landmarks: lateral and medial malleoli, lateral and medial knee joint lines, and greater trochanters.
value respectively between heel strike and midstance (50% between heel contact and toe-off). Kinematic data in the frontal plane were analyzed at first peak KAM while sagittal plane kinematic data were analyzed at peak knee flexion moment. Rate of change of net hip and knee joint force was calculated as the maximum slope of the resultant hip and knee force vs. time curve during the first 10% of stance. Although net resultant joint reaction force is not the same and should not be conflated with joint contact force (Vigotsky et al., 2019), we chose to include this measure in order to compare the effects of gait modifications in our sample to rate change of joint forces in a previous study implementing ‘‘trunk sway” gait in healthy individuals (Mündermann et al., 2008).
2.6. Data analysis Analyses were performed using SPSS (Version 24, IBM Corp, Armonk, NY) with alpha level set a priori at 0.05. Descriptive information were obtained using means and standard deviations. Repeated measures analysis of variance was used to determine if differences existed between dependent variables among conditions. Mauchly’s Test of Sphericity was conducted. When sphericity
was not met, Greenhouse-Geisser correction was used. Where results were significant, pairwise comparisons with Bonferroni adjustments were used to evaluate which conditions were significantly different from baseline.
3. Results All modifications decreased KAM (FP = 7%, MKT = 41%, TL = 5%), however, only MKT resulted in a significant reduction from baseline (p < 0.001) (Fig. 3). Changes in kinematic and temporospatial variables are shown in Table 1. Compared to baseline, participants decreased knee angle by 1.18° (p = 0.013), increased foot angle by 8.02° (p < 0.001), and increased trunk angle by 3.94° (p < 0.001) compared to baseline during MKT, FP, and TL modifications respectively. Stride width increased from baseline during TL (p = 0.023). Hip flexion angle was increased from baseline during all modifications (p < 0.001), while hip adduction angle was only significantly decreased from baseline during TL (p < 0.001). A significant main effect was found for hip abduction moment across modifications (F1.65,31.38 = 6.306, p = 0.007, g2 = 0.249) (Fig. 2). Both MKT and TL resulted in significant reductions in hip abduction moment from baseline of 23% (p < 0.001) and 14%
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
4
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
Fig. 2. Feedback given to participants during gait modification trials. The vertical axis represents the current modification parameter, while the horizontal axis is the percent of stance. The middle bandwidth represents the target range, with the lower boundary representing 3 SD greater (TL and FP) or lesser (MKT) than baseline and the upper boundary representing 5 SD greater (TL and FP) or lesser (MKT) than baseline. The scrolling line displayed the angle of the current modification parameter over stance. The line was updated in real-time during each step taken on the force plates. The trial was deemed successful if the red line was judged to be within the bandwidth between approximately 20–40% of stance.
(p = 0.002) respectively. Hip abduction moment was not significantly different from baseline during FP (p = 0.08) despite a 6% reduction. Hip extension moment was unchanged for all modifications (F3,57 = 1.875, p = 0.144, g2 = 0.090). A significant main effect was also found for rate change of hip net joint reaction force (F3,57 = 11.046, p < 0.001, g2 = 0.368), however, this was only significantly different from baseline during TL (p < 0.001). Similarly, a significant main effect was found for rate change of knee net joint reaction force (F3,57 = 12.599, p = <0.001, g2 = 0.399). Again, this was only significantly different from baseline during TL (p < 0.001) (Fig. 4).
4. Discussion The purpose of this study was to investigate how FP, MKT, and TL modifications designed to reduce KAM affect hip abduction moment. As expected, all modifications altered their target parameter in the intended direction. Reduction in knee (Barrios et al., 2010; Ferrigno et al., 2016; Gerbrands et al., 2017) and increase in foot progression angle (Shull et al., 2013a; Shull et al., 2013b) were comparable to those seen in previous studies, however, increase in trunk angle was smaller than prior studies (Hunt el al., 2011; Simic et al., 2012; Mündermann et al., 2008). Medial knee thrust reduced KAM by an average of 41%, similar to reductions between 20% and 38% seen previously (Barrios et al., 2010; Ferrigno et al., 2016; Gerbrands et al., 2017). Foot progression and TL also reduced KAM by 7% and 5%, respectively, though these reductions were not significant. The average KAM reduction seen from FP was slightly lower than reported in prior studies (Shull et al., 2013a; Shull et al., 2013b). Reductions in KAM seen during TL were similar to previous studies that reported a decrease of 7% with 4° of lean (Hunt et al., 2011) compared to our 5% with the same magnitude of lean. Larger magnitudes of TL have pro-
duced larger decreases in KAM of 21% and 25% with leans of 8° and 12° respectively (Hunt et al., 2011), suggesting that the small magnitude of trunk lean achieved in this study may have limited KAM reductions. Our hypothesis that all modifications would reduce hip abduction moment compared to baseline was partially supported, as only MKT and TL resulted in significant reductions of 23% and 14% respectively. Reduced foot progression resulted in a decrease of hip abduction moment of 6% yet did not reach statistical significance. Few studies have measured change in hip abduction moment after gait modification, however, Mündermann et al. (2008) reported a 55.3% reduction in hip abduction moment when walking with increased trunk sway (Mündermann et al., 2008). The smaller reduction in hip moment found in the current study may be explained by smaller average trunk lean (4°) compared to the prior (10°). The results of this study suggest that MKT also significantly reduces hip abduction moment; however, the two modifications appear to use different strategies. Trunk lean is designed to shift COM towards the symptomatic limb, lateralizing the GRF vector, and shortening the moment arm (Hunt et al., 2010; Mündermann et al., 2008; Simic et al., 2011). In addition, the hip was significantly more adducted during TL, meaning that similar to the knee joint during MKT, the hip joint is brought medially compared to normal gait, reducing the frontal plane moment arm further. Step width was also significantly increased with TL, which is noteworthy as increased step width has been a successful intervention to reduce KAM (Brindle et al., 2014; Paquette et al., 2015; Richards et al., 2018). The combination of lateralized GRF vector, a more medial hip joint, and increased step width likely contribute to the large reduction in hip abduction moment with TL. In comparison, MKT not only alters the center of pressure, lateralizing the GRF vector (Shull et al., 2013a) but also brings the knee joint medially which shortens the moment arm (Richards et al., 2018). This may explain why MKT reduced KAM
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
5
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
Fig. 3. Percent change in KAM (Nm/Kgm) and hip abduction moment (Nm/Kgm) during stance across gait modifications. Significance level of p < 0.05 denoted by * for MKT and y for TL.
Table 1 Descriptive and inferential statistics for kinematic, gait parameter, and temporospatial variables across gait modifications. Mean (SD) values, as well as effect sizes (Partial Eta2) are provided for each gait modification. Frontal and sagittal plane hip angles are reported at peak knee abduction moment. Gait parameters are reported as the mean across 100% stance. Significance level of p < 0.05 is denoted by *. Main Effect
Gait Modification Foot Progression
Medial Knee Thrust
Trunk Lean
F-val
p-val
Partial Eta2
Mean (SD)
Mean (SD)
Mean (SD)
Mean (SD)
Kinematics Frontal Plane Hip Angle (°) Sagittal Plane Hip Angle (°)
6.306 12.83
0.007 <0.001
0.249 0.403
4.85 (4.27) 12.94 (8.87)
4.76 (3.95) 15.47 (9.13)*
3.78 (4.94) 18.18 (10.22)*
2.49 (4.93)* 16.24 (10.14)*
Gait Parameters Foot Progression Angle (°) Knee Angle (°) Trunk Angle (°)
20.62 10.91 38.30
<0.001 0.001 <0.001
0.520 0.365 0.668
4.12 (4.40) 3.01 (2.53) 2.03 (2.26)
3.90 (5.31)* 2.58 (2.51)* 1.82 (2.26)
1.01 (10.15)* 1.83 (2.52)* 3.02 (3.12)*
4.44 (0.09) 2.89 (2.73) 5.97 (1.98)*
Temporospatial Speed (m/s) Stride Width (m) Stride Length (m)
1.03 3.54 0.96
0.387 0.020 0.419
0.051 0.157 0.048
1.37 (0.18) 0.13 (0.03) 1.44 (0.18)
1.36 (0.19) 0.14 (0.03) 1.42 (0.18)
1.36 (0.19) 0.14 (0.04) 1.45 (0.17)
1.36 (0.20) 0.15 (0.04)* 1.43 (0.16)
by 41%, whereas TL reduced KAM values only 5%. However, the medial movement of the knee does not explain the discrepancy in hip abduction moment reductions. This may be explained, at least partially, by smaller increases in hip adduction during MKT compared to TL. Therefore, although both modifications result in a more lateral GRF vector and adducted hip, reducing hip abduction moment, these changes appear greater in TL compared to MKT.
Baseline
Prior evidence suggests that lower hip abduction moment is predictive of KOA progression, possibly related to the purported relationship between hip abduction moment and hip abductor strength (Chang et al., 2005). It has been hypothesized that KOA patients develop weaker hip abductor musculature due to reduced demand at the hip as they naturally adopt compensatory gait strategies similar to TL to reduce frontal plane knee loads. The natural byproduct of which (as supported by the results of this study)
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
6
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
Fig. 4. Rate change of joint reaction force (%BW/s) for the hip and knee across gait modifications. Significance level of p < 0.05 is denoted by *.
is reduced hip abduction moment (Mündermann et al., 2005). In turn, weaker hip musculature may reduce their ability to regulate knee joint loads. The most current research, however, finds little evidence to support a relationship between hip abduction moment and strength (Kean et al., 2015; Rutherford and Hubley-Kozey, 2009). Although the link between hip abduction moment and strength still requires clarification, additional evidence suggests that higher daily cumulative hip abduction moment is predictive of the progression of hip osteoarthritis (Tateuchi et al., 2017). After 12 months, higher cumulative daily hip abduction moment resulted in a cartilage thickness loss of >0.5 mm (Tateuchi et al., 2017). This suggests that gait modifications such as MKT and TL may be appropriate for individuals with hip osteoarthritis, and/or reduce the likelihood of secondary hip OA progression. Trunk lean increased rate change of both hip and knee joint reaction forces compared to baseline. Mündermann et al. (2008) investigated the effect of trunk sway on hip and knee joint reaction force loading rates in healthy individuals and found no differences in rates at either joint. The difference is likely explained by methodological disparities. Mündermann et al. employed a ‘trunk sway’ modification which, although similar to TL, is executed differently. Whereas the instructions for participants using trunk sway in Mündermann’s study were ‘‘move your trunk more from side to side”, we instructed participants to ‘‘lean your trunk towards your dominant limb immediately after your foot makes contact with the ground”, and to ‘‘imagine bringing your whole upper body over the dominant leg, rather than just laterally flexing your trunk”. The TL modification employed in this study mimics the natural compensatory gait often adopted by KOA patients to offset medial compartment loads where ipsilateral trunk lean is increased over the symptomatic limb during stance (Briem and Snyder-Mackler, 2009; Hunt et al., 2010) and the knee is in a more extended position at heel strike (Mündermann et al., 2005). When adopting TL, participants in our study presented a more extended hip (FP: 16.5° MKT: 19.9° TL: 15.6°) and knee (FP: 6.8° MKT: 12.0° TL: 4.8°) compared to the other modifications over the first 10% of stance, whereas knee flexion was actually increased using increased trunk sway in the prior study. Patients using this type of naturally adopted ‘trunk lean’ strategy have previously been shown to have greater rate changes in joint reaction forces compared to healthy controls (Mündermann et al., 2005). Increased rate change in joint reaction force has been hypothesized to increase cartilage damage and increase the risk of developing secondary hip osteoarthritis (Mündermann et al., 2005), however, due to the fact that these are resultant joint forces and not bone contact forces, caution should be taken when interpreting these results. Resultant joint forces may not be an accurate representa-
tion of bone contact forces as they do not include the forces produced by internal tissues (Vigotsky et al., 2019). Therefore, it may be inappropriate to infer that an increase in resultant net joint force loading rate may lead to adverse effects at the cartilage and bone level. If possible, future studies should consider examining whether the rate of change of bone contact force increases with GMs, as an increased bone loading rate has been associated with increased cartilage damage (Ewers et al., 2002). Our sample was composed of healthy individuals, and although this limits the application of our results, they provide an indication of how KOA patients may respond to the same interventions. We did not randomize the order of modification introducing the possibility of order effects. For example, it is possible that by performing MKT after TL, participants demonstrated increased hip adduction during MKT that they would not have otherwise. Instead, we chose to progress participants from the most difficult to the most easily adopted modification. This was based on pilot data collection sessions where we found that participants often struggled to achieve the prescribed modification near the end of a lengthy data collection session due to mental and/or physical fatigue. We used a novel method of visual RTB where we prescribed individualized target ranges calculated from each participant’s baseline gait. The advantage of this method was that it standardized the amount of kinematic change across participants. This range, however, was greatly affected by variation in participants’ baseline gait. Individuals with smaller SD at baseline had narrower target ranges, making it more difficult to stay within the prescribed range. Due to the small target range created by our standardization technique, a subjective measure of trial success was used that may have introduced variability in the achieved kinematic changes. This study demonstrates that gait modifications designed to reduce KAM, such as MKT and TL, subsequently reduce hip abduction moment. Clinical use of gait modification should consider multiple factors including KAM, hip abduction moment, and subjective measures such as ease of adoption. While MKT resulted in the largest KAM reduction, anecdotally it was much more difficult to adopt (along with TL) compared to FP. Because of this, MKT may not represent a preferred gait modification for most patients despite its large effect. Likewise, the decrease in hip abduction moment found during TL and MKT may also limit their clinical efficacy. Additional research is needed to fully understand the effect of gait modifications in clinical populations, particularly the relationship between hip abduction moments and KOA progression. Funding None. Declaration of Competing Interest None of the authors have any financial or personal relationships that could inappropriately influence this manuscript. References Amin, S., Luepongsak, N., McGibbon, C.A., LaValley, M.P., Krebs, D.E., Felson, D.T., 2004. Knee adduction moment and development of chronic knee pain in elders. Arthritis Care Res. 51, 371–376. Astephen, J.L., Deluzio, K.J., Caldwell, G.E., Dunbar, M.J., 2008. Biomechanical changes at the hip, knee, and ankle joints during gait are associated with knee osteoarthritis severity. J. Orthop. Res. 26, 332–341. Barrios, J.A., Crossley, K.M., Davis, I.S., 2010. Gait retraining to reduce the knee adduction moment through real-time visual feedback of dynamic knee alignment. J. Biomech. 43, 2208–2213. Briem, K., Snyder-Mackler, L., 2009. Proximal gait adaptations in medial knee OA. J. Orthop. Res. 27, 78–83. Brindle, R.A., Milner, C.E., Zhang, S., Fitzhugh, E.C., 2014. Changing step width alters lower extremity biomechanics during running. Gait Posture 39, 124–128.
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx Butler, R.J., Minick, K.I., Ferber, R., Underwood, F., 2009. Gait mechanics after ACL reconstruction: implications for the early onset of knee osteoarthritis. Br. J. Sports Med. 43, 366–370. Chang, A., Hayes, K., Dunlop, D., Song, J., Hurwitz, D., Cahue, S., Sharma, L., 2005. Hip abduction moment and protection against medial tibiofemoral osteoarthritis progression. Arthritis Rheum. 52, 3515–3519. Cortes, N., Quammen, D., Lucci, S., Greska, E., Onate, J., 2012. A functional agility short-term fatigue protocol changes lower extremity mechanics. J. Sports Sci. 30, 797–805. Eddo, O., Lindsey, B., Caswell, S.V., Cortes, N., 2017. Current evidence of gait modification with real-time biofeedback to alter kinetic, temporospatial, and function-related outcomes: a review. Int. J. Kinesiol. Sports Sci. 5, 35–55. Ewers, B.J., Jayaraman, V.M., Banglmaier, R.F., Haut, R.C., 2002. Rate of blunt impact loading affects changes in retropatellar cartilage and underlying bone in the rabbit patella. J. Biomech. 35, 747–755. Ferrigno, C., Stoller, I.S., Shakoor, N., Thorp, L.E., Wimmer, M.A., 2016. The feasibility of using augmented auditory feedback from a pressure detecting insole to reduce the knee adduction moment: a proof of concept study. J. Biomech. Eng. 138, 021014-1–021014-7. Gerbrands, T.A., Pisters, M.F., Theeven, P.J.R., Verschueren, S., Vanwanseele, B., 2017. Lateral trunk lean and medializing the knee as gait strategies for knee osteoarthritis. Gait Posture 51, 247–253. Grood, E.S., Suntay, W.J., 1983. A joint coordinate system for the clinical description of three-dimensional motions: application the knee. J. Biomech. Eng. 105, 136– 144. Hunt, M.A., Simic, M., Hinman, R.S., Bennell, K.L., Wrigley, T.V., 2011. Feasibility of a gait retraining strategy for reducing knee joint loading: increased trunk lean guided by real-time biofeedback. J. Biomech. 44, 943–947. Hunt, M.A., Wrigley, T.V., Hinman, R.S., Bennell, K.L., 2010. Individuals with severe knee osteoarthritis (OA) exhibit altered proximal walking mechanics compared with individuals with less severe OA and those without knee pain. Arthritis Care Res. 62, 1426–1432. Hunt, M., Birmingham, T., Bryant, D., Jones, I., Giffin, J., Jenkyn, T., Vandervoort, A., 2008. Lateral trunk lean explains variation in dynamic knee joint load in patients with medial compartment knee osteoarthritis. Osteoarth. Cartil. 16, 591–599. Kean, C.O., Bennell, K.L., Wrigley, T.V., Hinman, R.S., 2015. Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis. Clin. Biomech. 30, 226–230. Miyazaki, T., Wada, M., Kawahara, H., Sato, M., Baba, H., Shimada, S., 2002. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann. Rheum. Dis. 61, 617–622. Mündermann, A., Asay, J.L., Mündermann, L., Andriacchi, T.P., 2008. Implications of increased medio-lateral trunk sway for ambulatory mechanics. J. Biomech. 41, 165–170.
7
Mündermann, A., Dyrby, C.O., Andriacchi, T.P., 2005. Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arthritis Rheum. 52, 2835–2844. Paquette, M.R., Klipple, G., Zhang, S., 2015. Greater step widths reduce internal knee abduction moments in medial compartment knee osteoarthritis patients during stair ascent. J. Appl. Biomech. 31, 229–236. Reeves, N.D., Bowling, F.L., 2011. Conservative biomechanical strategies for knee osteoarthritis. Nat. Rev. Rheumatol. 7, 113. Richards, R.E., van den Noort, J.C., van der Esch, M., Booij, M.J., Harlaar, J., 2018. Effect of real-time biofeedback on peak knee adduction moment in patients with medial knee osteoarthritis: is direct feedback effective?. Clin. Biomech. 57, 150–158. Richards, R., van den Noort, J.C., Dekker, J., Harlaar, J., 2017. Gait retraining with real-time biofeedback to reduce knee adduction moment: systematic review of effects and methods used. Arch. Phys. Med. Rehabil. 98, 137–150. Rutherford, D.J., Hubley-Kozey, C., 2009. Explaining the hip adduction moment variability during gait: Implications for hip abductor strengthening. Clin. Biomech. 24, 267–273. Schwartz, M.H., Rozumalski, A., 2005. A new method for estimating joint parameters from motion data. J. Biomech. 38, 107–116. Sharma, L., Hurwitz, D.E., Thonar, E.J.-M.A., Sum, J.A., Lenz, M.E., Dunlop, D.D., Schnitzer, T.J., Kirwan-Mellis, G., Andriacchi, T.P., 1998. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum. 41, 1233–1240. Shull, P.B., Shultz, R., Silder, A., Dragoo, J.L., Besier, T.F., Cutkosky, M.R., Delp, S.L., 2013a. Toe-in gait reduces the first peak knee adduction moment in patients with medial compartment knee osteoarthritis. J. Biomech. 46, 122–128. Shull, P.B., Silder, A., Shultz, R., Dragoo, J.L., Besier, T.F., Delp, S.L., Cutkosky, M.R., 2013b. Six-week gait retraining program reduces knee adduction moment, reduces pain, and improves function for individuals with medial compartment knee osteoarthritis. J. Orthop. Res. 31, 1020–1025. Simic, M., Hinman, R.S., Wrigley, T.V., Bennell, K.L., Hunt, M.A., 2011. Gait modification strategies for altering medial knee joint load: a systematic review. Arthritis Care Res. 63, 405–426. Simic, M., Hunt, M.A., Bennell, K.L., Hinman, R.S., Wrigley, T.V., 2012. Trunk lean gait modification and knee joint load in people with medial knee osteoarthritis: the effect of varying trunk lean angles. Arthritis Care Res. 64, 1545–1553. Tateuchi, H., Koyama, Y., Akiyama, H., Goto, K., So, K., Kuroda, Y., Ichihashi, N., 2017. Daily cumulative hip moment is associated with radiographic progression of secondary hip osteoarthritis. Osteoarth. Cartil. 25, 1291–1298. Vigotsky, Andrew D., Zelik, K.E., Lake, J., Hinrichs, R.N., 2019. Mechanical misconceptions: have we lost the ‘‘mechanics” in ‘‘sports biomechanics”?. J. Biomech. 93, 1–5. Winter, D., 2009. Biomechanics and Motor Control of Human Movement. John Wiley & Sons Inc, Hoboken.
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
Contents lists available at ScienceDirect
Journal of Biomechanics journal homepage: www.elsevier.com/locate/jbiomech www.JBiomech.com
Changes in hip mechanics during gait modification to reduce knee abduction moment Sarah Kettlety a, Bryndan Lindsey a, Oladipo Eddo a, Matt Prebble a, Shane Caswell a, Nelson Cortes a,b,⇑ a b
Sports Medicine Assessment, Research & Testing (SMART) Laboratory, George Mason University, School of Kinesiology, Manassas, VA, United States Department of Bioengineering, George Mason University, Fairfax, VA, United States
a r t i c l e
i n f o
Article history: Accepted 12 November 2019 Available online xxxx Keywords: Knee osteoarthritis Gait retraining Real-time biofeedback Hip abduction moment
a b s t r a c t First peak knee abduction moment (KAM) has been associated with the severity and progression of knee osteoarthritis (KOA). Gait modifications, including lateral trunk lean (TL), medial knee thrust (MKT), and reduced foot progression (FP) have decreased KAM. However, their effects on the hip joint are poorly understood. Reduced hip abduction moment has been found to be predictive of KOA progression and has been hypothesized to represent a decreased demand on the hip musculature. Lack of studies investigating changes in hip mechanics as a result of gait modification limits our understanding of their cumulative benefit, therefore, we investigated the effects of TL, MKT, and FP on internal hip abduction moment as well as rate change in net joint reaction force. Using real-time visual biofeedback, five trials were completed for each modification. Each modification target range was individualized to 3–5 SD greater (TL and FP) or lesser (MKT) than the participants mean baseline value. Kinematics and kinetics at the hip and knee were calculated at first peak KAM. Trunk lean and MKT decreased hip abduction moment compared to baseline (p < 0.001). Trunk lean increased rate change in net joint reaction force at both the hip (p < 0.001) and knee (p < 0.001) compared to baseline. Additional research is needed to fully understand the effect of gait modifications in a clinical population, particularly the relationship between hip abduction moments and KOA progression. Although interventions such as MKT and TL can be successful in reducing KAM, their effects on hip abduction moment should be considered before clinical implementation. Ó 2019 Elsevier Ltd. All rights reserved.
1. Introduction The severity (Sharma et al., 1998), symptoms (Amin et al., 2004), and progression (Miyazaki et al., 2002) of knee osteoarthritis (KOA) have been related to increased frontal plane knee moment. The first peak internal knee abduction moment (KAM) is a commonly reported surrogate measure of joint loading (Eddo et al., 2017; Richards et al., 2017). Gait modifications utilizing real-time biofeedback (RTB) have been effective in reducing KAM in both healthy and KOA populations (Eddo et al., 2017; Richards et al., 2017). For example, medial knee thrust (MKT), lateral trunk lean (TL), and reduced foot progression (FP) have shown reductions in KAM of up to 38% (Ferrigno et al., 2016), 65% (Mündermann et al., 2008), and 20% (Shull et al., 2013b) respectively.
⇑ Corresponding author at: Sports Medicine Assessment, Research & Testing (SMART) Laboratory, George Mason University, 10890 George Mason Circle, KJH 201E, MSN 4E5 Manassas, VA 20110, United States. E-mail address: [email protected] (N. Cortes).
Although these modifications have shown themselves able to reduce KAM, their effects on other lower extremity joints remain unclear (Simic et al., 2011). Patients with KOA have lower hip abduction moments compared to healthy counterparts (Astephen et al., 2008; Briem & Snyder-Mackler, 2009) with more severe KOA corresponding to lower hip abduction moments (Astephen et al., 2008; Mündermann et al., 2005). These results suggest that other mechanical measures, such as hip abduction moment, are also important in addition to KAM. The goal of gait modification is to decrease pain, improve function, and slow the progression of KOA. This can be achieved by various mechanisms including shifting the center of mass (COM) towards the symptomatic knee with TL (Hunt et al., 2010; Mündermann et al., 2008; Simic et al., 2011), or shifting of the center of pressure laterally (Shull et al., 2013a) with concomitant medialization of the symptomatic knee joint (Reeves and Bowling, 2011) seen during FP and MKT. Both mechanisms have the effect of lateralizing the GRF vector, reducing frontal plane moment arm and subsequent knee abduction moment. Lateralization of the GRF vector not only affects the knee joint, as limited
https://doi.org/10.1016/j.jbiomech.2019.109509 0021-9290/Ó 2019 Elsevier Ltd. All rights reserved.
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
2
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
evidence has demonstrated an association between decreased KAM and reduced hip abduction moment (Astephen et al., 2008; Mündermann et al., 2005). Lack of studies investigating changes in hip mechanics with gait modification limits our understanding of their cumulative benefit. The available literature suggests that TL significantly decreases hip abduction moment (Briem & Snyder-Mackler, 2009; Mündermann et al., 2008), and increases the rate change of net joint reaction forces (Mündermann et al., 2005). The effects of other commonly studied gait modifications remain unknown. Therefore, the purpose of this study was to investigate how FP, MKT, and TL gait modifications affect internal hip abduction moment. We hypothesized that all gait modifications would significantly reduce KAM and hip abduction moment from baseline.
mean and standard deviation (SD) for three modification parameters: frontal plane trunk and knee angle, and transverse plane foot angle. Trunk angle was defined as the frontal plane deviation of the trunk segment represented by the right scapula, 10th thoracic, and left/right lower back markers from the vertical laboratory axis (Hunt et al., 2011). Knee angle was defined as the frontal plane knee angle created between the thigh and shank segments (Barrios et al., 2010). Foot angle was found as the offset between the lines formed by the posterior calcaneus and 2nd metatarsophalangeal joint markers, and the anterior-posterior laboratory axis (Shull et al., 2013a). Increased trunk lean to the dominant limb was quantified as positive, increased knee abduction as negative, and reduced foot progression angle as positive. 2.4. Gait modification trials
2. Methods 2.1. Participants Twenty healthy participants (26.7 ± 4.7 years, 1.75 ± 0.1 m, 73.4 ± 12.4 kg) volunteered for this study after giving informed consent approved by the Institution Review Board. A within-group single-session repeated measures design was used to compare joint kinematics and kinetics of participants’ dominant limb across gait conditions. Dominant limb was defined as the preferred leg in a kicking task (Cortes et al., 2012). Eighteen participants were right leg dominant, two left. Participants were only eligible if they were free from any knee, hip, or back pain that required treatment within the prior 6 months and no history of lower limb or back surgery. Exclusion criteria included any neurological or musculoskeletal impairment that would affect gait or any cognitive impairment that would inhibit motor learning. 2.2. Instrumentation Fifty three retroreflective markers were attached to the trunk and lower extremities of participants (Fig. 1). Eight high-speed motion analysis cameras (Vicon, Oxford, England) sampling at 200 Hz tracked marker trajectory. Ground reaction force was acquired using four floor embedded force plates aligned in a 2.4 m long row sampling at 1000 Hz (Bertec, Columbus, OH). A static calibration trial was collected by having participants stand on a force plate with both feet parallel to the anterior-posterior axis of the laboratory. Participants also performed a dynamic calibration to estimate hip joint center by completing three clockwise rotations of the pelvis (Schwartz and Rozumalski, 2005). Calibration makers were removed for walking trials. From the static trial, a kinematic model was created for each participant using Visual 3D software (C-Motion, Germantown MD, USA) which included the trunk, pelvis, and bilateral thigh, shank, and foot segments. 2.3. Baseline trials For all trials, participants walked along a 6-meter long walkway aligned parallel with the anterior – posterior axis of the laboratory at a self-selected speed. Tape was placed at the lateral borders of the walkway in line with the edges of the force plates. Participants were instructed to stay within the tape during trials to prevent deviation from a straight walking path. At the end of each trial, participants returned to the beginning of the walkway, starting at the same point as the prior trial. Timing gates (Brower Timing Systems, Draper UT, USA) positioned at each end of the force plates recorded walking speed during 10 baseline trials. For a trial to be successful, one full contact on a force plate with the dominant foot was required. Recorded data were exported to Visual 3D to calculate
Five trials were completed for each modification using visual RTB, which was delivered in the form of a line graph projected on the wall in front of the walkway (Fig. 2). The graph displayed the angle of the current modification parameter over stance and was updated after each step of the dominant limb. A bandwidth representing a range of 3–5 SD greater (FP and TL) or lesser (MKT) than baseline was displayed on the graph. Participants were instructed to walk so that the line representing the gait parameter fell within the bandwidth. Modification trials were completed in the following order: TL, MKT, and FP. Participants were provided standardized verbal instructions before implementing each modification and completed as many practice trials as needed to familiarize themselves with each modification. Additional verbal feedback was provided during practice trials as needed. Successful trials required one full foot contact of the dominant limb within a force plate, average gait speed of ±5% relative to baseline, and that the line representing the target modification parameter fell within the highlighted bandwidth between approximately 20% to 40% of stance. This last measure of trial success was chosen as some participants found it difficult to stay within the bandwidth during the entire stance period during pilot trials. As we were interested in reducing the first peak of the adduction moment, we counted trials as successful if participants could modify their kinematics at approximately 30% of stance, where peak KAM typically occurs (Hunt et al., 2008; Simic et al., 2011). Only successful trials counted towards the 5 required for each modification. 2.5. Data processing The kinematic model created in Visual 3D was used to quantify the motion at the hip, knee, and ankle joints with rotations being expressed relative to the static trial. A cardan angle sequence was used to calculate joint angles (Grood and Suntay, 1983) and standard inverse dynamics analysis was conducted to synthesize the trajectory and vertical ground reaction force data for internal joint moment estimation at the knee and hip. Although external joint moments are most commonly reported in KOA literature, internal moments resist the action of external moments and can be thought of as equal but opposite in sign (Butler et al., 2009). Joint kinematics and kinetics were smoothed using a low-pass Butterworth filter with a cut off frequency of 8 Hz to reduce the effects of artifacts based on results from residual analysis (Winter, 2009). Joint angles were measured in degrees, and internal joint moments were normalized to mass and height (Nm/Kgm). Ground reaction force data were normalized to body weight and gait trials were normalized to 100 percent of stance. Modification parameters were calculated as the average across stance. First peak for knee and hip abduction moment and hip flexion moment were defined as the peak minimum and maximum
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
3
Fig. 1. Experimental marker placement. Four tracking clusters (18 markers) were placed on the lateral aspect of each thigh and shank. Additional tracking markers (22 markers) were attached to the manubrium, 7th cervical vertebrae, right scapula, 10th thoracic vertebrae, and bilaterally to the following locations: posterior and lateral calcaneus, 5th distal metatarsal, 1st proximal metatarsal, 2nd metatarsophalangeal joint, tibial tuberosity, lateral iliac spine, posterior superior iliac spine, and acromion. Three tracking markers, arranged to form a triangular cluster, were attached to the lower back. Ten additional calibration markers were attached bilaterally to the following anatomical landmarks: lateral and medial malleoli, lateral and medial knee joint lines, and greater trochanters.
value respectively between heel strike and midstance (50% between heel contact and toe-off). Kinematic data in the frontal plane were analyzed at first peak KAM while sagittal plane kinematic data were analyzed at peak knee flexion moment. Rate of change of net hip and knee joint force was calculated as the maximum slope of the resultant hip and knee force vs. time curve during the first 10% of stance. Although net resultant joint reaction force is not the same and should not be conflated with joint contact force (Vigotsky et al., 2019), we chose to include this measure in order to compare the effects of gait modifications in our sample to rate change of joint forces in a previous study implementing ‘‘trunk sway” gait in healthy individuals (Mündermann et al., 2008).
2.6. Data analysis Analyses were performed using SPSS (Version 24, IBM Corp, Armonk, NY) with alpha level set a priori at 0.05. Descriptive information were obtained using means and standard deviations. Repeated measures analysis of variance was used to determine if differences existed between dependent variables among conditions. Mauchly’s Test of Sphericity was conducted. When sphericity
was not met, Greenhouse-Geisser correction was used. Where results were significant, pairwise comparisons with Bonferroni adjustments were used to evaluate which conditions were significantly different from baseline.
3. Results All modifications decreased KAM (FP = 7%, MKT = 41%, TL = 5%), however, only MKT resulted in a significant reduction from baseline (p < 0.001) (Fig. 3). Changes in kinematic and temporospatial variables are shown in Table 1. Compared to baseline, participants decreased knee angle by 1.18° (p = 0.013), increased foot angle by 8.02° (p < 0.001), and increased trunk angle by 3.94° (p < 0.001) compared to baseline during MKT, FP, and TL modifications respectively. Stride width increased from baseline during TL (p = 0.023). Hip flexion angle was increased from baseline during all modifications (p < 0.001), while hip adduction angle was only significantly decreased from baseline during TL (p < 0.001). A significant main effect was found for hip abduction moment across modifications (F1.65,31.38 = 6.306, p = 0.007, g2 = 0.249) (Fig. 2). Both MKT and TL resulted in significant reductions in hip abduction moment from baseline of 23% (p < 0.001) and 14%
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
4
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
Fig. 2. Feedback given to participants during gait modification trials. The vertical axis represents the current modification parameter, while the horizontal axis is the percent of stance. The middle bandwidth represents the target range, with the lower boundary representing 3 SD greater (TL and FP) or lesser (MKT) than baseline and the upper boundary representing 5 SD greater (TL and FP) or lesser (MKT) than baseline. The scrolling line displayed the angle of the current modification parameter over stance. The line was updated in real-time during each step taken on the force plates. The trial was deemed successful if the red line was judged to be within the bandwidth between approximately 20–40% of stance.
(p = 0.002) respectively. Hip abduction moment was not significantly different from baseline during FP (p = 0.08) despite a 6% reduction. Hip extension moment was unchanged for all modifications (F3,57 = 1.875, p = 0.144, g2 = 0.090). A significant main effect was also found for rate change of hip net joint reaction force (F3,57 = 11.046, p < 0.001, g2 = 0.368), however, this was only significantly different from baseline during TL (p < 0.001). Similarly, a significant main effect was found for rate change of knee net joint reaction force (F3,57 = 12.599, p = <0.001, g2 = 0.399). Again, this was only significantly different from baseline during TL (p < 0.001) (Fig. 4).
4. Discussion The purpose of this study was to investigate how FP, MKT, and TL modifications designed to reduce KAM affect hip abduction moment. As expected, all modifications altered their target parameter in the intended direction. Reduction in knee (Barrios et al., 2010; Ferrigno et al., 2016; Gerbrands et al., 2017) and increase in foot progression angle (Shull et al., 2013a; Shull et al., 2013b) were comparable to those seen in previous studies, however, increase in trunk angle was smaller than prior studies (Hunt el al., 2011; Simic et al., 2012; Mündermann et al., 2008). Medial knee thrust reduced KAM by an average of 41%, similar to reductions between 20% and 38% seen previously (Barrios et al., 2010; Ferrigno et al., 2016; Gerbrands et al., 2017). Foot progression and TL also reduced KAM by 7% and 5%, respectively, though these reductions were not significant. The average KAM reduction seen from FP was slightly lower than reported in prior studies (Shull et al., 2013a; Shull et al., 2013b). Reductions in KAM seen during TL were similar to previous studies that reported a decrease of 7% with 4° of lean (Hunt et al., 2011) compared to our 5% with the same magnitude of lean. Larger magnitudes of TL have pro-
duced larger decreases in KAM of 21% and 25% with leans of 8° and 12° respectively (Hunt et al., 2011), suggesting that the small magnitude of trunk lean achieved in this study may have limited KAM reductions. Our hypothesis that all modifications would reduce hip abduction moment compared to baseline was partially supported, as only MKT and TL resulted in significant reductions of 23% and 14% respectively. Reduced foot progression resulted in a decrease of hip abduction moment of 6% yet did not reach statistical significance. Few studies have measured change in hip abduction moment after gait modification, however, Mündermann et al. (2008) reported a 55.3% reduction in hip abduction moment when walking with increased trunk sway (Mündermann et al., 2008). The smaller reduction in hip moment found in the current study may be explained by smaller average trunk lean (4°) compared to the prior (10°). The results of this study suggest that MKT also significantly reduces hip abduction moment; however, the two modifications appear to use different strategies. Trunk lean is designed to shift COM towards the symptomatic limb, lateralizing the GRF vector, and shortening the moment arm (Hunt et al., 2010; Mündermann et al., 2008; Simic et al., 2011). In addition, the hip was significantly more adducted during TL, meaning that similar to the knee joint during MKT, the hip joint is brought medially compared to normal gait, reducing the frontal plane moment arm further. Step width was also significantly increased with TL, which is noteworthy as increased step width has been a successful intervention to reduce KAM (Brindle et al., 2014; Paquette et al., 2015; Richards et al., 2018). The combination of lateralized GRF vector, a more medial hip joint, and increased step width likely contribute to the large reduction in hip abduction moment with TL. In comparison, MKT not only alters the center of pressure, lateralizing the GRF vector (Shull et al., 2013a) but also brings the knee joint medially which shortens the moment arm (Richards et al., 2018). This may explain why MKT reduced KAM
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
5
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
Fig. 3. Percent change in KAM (Nm/Kgm) and hip abduction moment (Nm/Kgm) during stance across gait modifications. Significance level of p < 0.05 denoted by * for MKT and y for TL.
Table 1 Descriptive and inferential statistics for kinematic, gait parameter, and temporospatial variables across gait modifications. Mean (SD) values, as well as effect sizes (Partial Eta2) are provided for each gait modification. Frontal and sagittal plane hip angles are reported at peak knee abduction moment. Gait parameters are reported as the mean across 100% stance. Significance level of p < 0.05 is denoted by *. Main Effect
Gait Modification Foot Progression
Medial Knee Thrust
Trunk Lean
F-val
p-val
Partial Eta2
Mean (SD)
Mean (SD)
Mean (SD)
Mean (SD)
Kinematics Frontal Plane Hip Angle (°) Sagittal Plane Hip Angle (°)
6.306 12.83
0.007 <0.001
0.249 0.403
4.85 (4.27) 12.94 (8.87)
4.76 (3.95) 15.47 (9.13)*
3.78 (4.94) 18.18 (10.22)*
2.49 (4.93)* 16.24 (10.14)*
Gait Parameters Foot Progression Angle (°) Knee Angle (°) Trunk Angle (°)
20.62 10.91 38.30
<0.001 0.001 <0.001
0.520 0.365 0.668
4.12 (4.40) 3.01 (2.53) 2.03 (2.26)
3.90 (5.31)* 2.58 (2.51)* 1.82 (2.26)
1.01 (10.15)* 1.83 (2.52)* 3.02 (3.12)*
4.44 (0.09) 2.89 (2.73) 5.97 (1.98)*
Temporospatial Speed (m/s) Stride Width (m) Stride Length (m)
1.03 3.54 0.96
0.387 0.020 0.419
0.051 0.157 0.048
1.37 (0.18) 0.13 (0.03) 1.44 (0.18)
1.36 (0.19) 0.14 (0.03) 1.42 (0.18)
1.36 (0.19) 0.14 (0.04) 1.45 (0.17)
1.36 (0.20) 0.15 (0.04)* 1.43 (0.16)
by 41%, whereas TL reduced KAM values only 5%. However, the medial movement of the knee does not explain the discrepancy in hip abduction moment reductions. This may be explained, at least partially, by smaller increases in hip adduction during MKT compared to TL. Therefore, although both modifications result in a more lateral GRF vector and adducted hip, reducing hip abduction moment, these changes appear greater in TL compared to MKT.
Baseline
Prior evidence suggests that lower hip abduction moment is predictive of KOA progression, possibly related to the purported relationship between hip abduction moment and hip abductor strength (Chang et al., 2005). It has been hypothesized that KOA patients develop weaker hip abductor musculature due to reduced demand at the hip as they naturally adopt compensatory gait strategies similar to TL to reduce frontal plane knee loads. The natural byproduct of which (as supported by the results of this study)
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
6
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx
Fig. 4. Rate change of joint reaction force (%BW/s) for the hip and knee across gait modifications. Significance level of p < 0.05 is denoted by *.
is reduced hip abduction moment (Mündermann et al., 2005). In turn, weaker hip musculature may reduce their ability to regulate knee joint loads. The most current research, however, finds little evidence to support a relationship between hip abduction moment and strength (Kean et al., 2015; Rutherford and Hubley-Kozey, 2009). Although the link between hip abduction moment and strength still requires clarification, additional evidence suggests that higher daily cumulative hip abduction moment is predictive of the progression of hip osteoarthritis (Tateuchi et al., 2017). After 12 months, higher cumulative daily hip abduction moment resulted in a cartilage thickness loss of >0.5 mm (Tateuchi et al., 2017). This suggests that gait modifications such as MKT and TL may be appropriate for individuals with hip osteoarthritis, and/or reduce the likelihood of secondary hip OA progression. Trunk lean increased rate change of both hip and knee joint reaction forces compared to baseline. Mündermann et al. (2008) investigated the effect of trunk sway on hip and knee joint reaction force loading rates in healthy individuals and found no differences in rates at either joint. The difference is likely explained by methodological disparities. Mündermann et al. employed a ‘trunk sway’ modification which, although similar to TL, is executed differently. Whereas the instructions for participants using trunk sway in Mündermann’s study were ‘‘move your trunk more from side to side”, we instructed participants to ‘‘lean your trunk towards your dominant limb immediately after your foot makes contact with the ground”, and to ‘‘imagine bringing your whole upper body over the dominant leg, rather than just laterally flexing your trunk”. The TL modification employed in this study mimics the natural compensatory gait often adopted by KOA patients to offset medial compartment loads where ipsilateral trunk lean is increased over the symptomatic limb during stance (Briem and Snyder-Mackler, 2009; Hunt et al., 2010) and the knee is in a more extended position at heel strike (Mündermann et al., 2005). When adopting TL, participants in our study presented a more extended hip (FP: 16.5° MKT: 19.9° TL: 15.6°) and knee (FP: 6.8° MKT: 12.0° TL: 4.8°) compared to the other modifications over the first 10% of stance, whereas knee flexion was actually increased using increased trunk sway in the prior study. Patients using this type of naturally adopted ‘trunk lean’ strategy have previously been shown to have greater rate changes in joint reaction forces compared to healthy controls (Mündermann et al., 2005). Increased rate change in joint reaction force has been hypothesized to increase cartilage damage and increase the risk of developing secondary hip osteoarthritis (Mündermann et al., 2005), however, due to the fact that these are resultant joint forces and not bone contact forces, caution should be taken when interpreting these results. Resultant joint forces may not be an accurate representa-
tion of bone contact forces as they do not include the forces produced by internal tissues (Vigotsky et al., 2019). Therefore, it may be inappropriate to infer that an increase in resultant net joint force loading rate may lead to adverse effects at the cartilage and bone level. If possible, future studies should consider examining whether the rate of change of bone contact force increases with GMs, as an increased bone loading rate has been associated with increased cartilage damage (Ewers et al., 2002). Our sample was composed of healthy individuals, and although this limits the application of our results, they provide an indication of how KOA patients may respond to the same interventions. We did not randomize the order of modification introducing the possibility of order effects. For example, it is possible that by performing MKT after TL, participants demonstrated increased hip adduction during MKT that they would not have otherwise. Instead, we chose to progress participants from the most difficult to the most easily adopted modification. This was based on pilot data collection sessions where we found that participants often struggled to achieve the prescribed modification near the end of a lengthy data collection session due to mental and/or physical fatigue. We used a novel method of visual RTB where we prescribed individualized target ranges calculated from each participant’s baseline gait. The advantage of this method was that it standardized the amount of kinematic change across participants. This range, however, was greatly affected by variation in participants’ baseline gait. Individuals with smaller SD at baseline had narrower target ranges, making it more difficult to stay within the prescribed range. Due to the small target range created by our standardization technique, a subjective measure of trial success was used that may have introduced variability in the achieved kinematic changes. This study demonstrates that gait modifications designed to reduce KAM, such as MKT and TL, subsequently reduce hip abduction moment. Clinical use of gait modification should consider multiple factors including KAM, hip abduction moment, and subjective measures such as ease of adoption. While MKT resulted in the largest KAM reduction, anecdotally it was much more difficult to adopt (along with TL) compared to FP. Because of this, MKT may not represent a preferred gait modification for most patients despite its large effect. Likewise, the decrease in hip abduction moment found during TL and MKT may also limit their clinical efficacy. Additional research is needed to fully understand the effect of gait modifications in clinical populations, particularly the relationship between hip abduction moments and KOA progression. Funding None. Declaration of Competing Interest None of the authors have any financial or personal relationships that could inappropriately influence this manuscript. References Amin, S., Luepongsak, N., McGibbon, C.A., LaValley, M.P., Krebs, D.E., Felson, D.T., 2004. Knee adduction moment and development of chronic knee pain in elders. Arthritis Care Res. 51, 371–376. Astephen, J.L., Deluzio, K.J., Caldwell, G.E., Dunbar, M.J., 2008. Biomechanical changes at the hip, knee, and ankle joints during gait are associated with knee osteoarthritis severity. J. Orthop. Res. 26, 332–341. Barrios, J.A., Crossley, K.M., Davis, I.S., 2010. Gait retraining to reduce the knee adduction moment through real-time visual feedback of dynamic knee alignment. J. Biomech. 43, 2208–2213. Briem, K., Snyder-Mackler, L., 2009. Proximal gait adaptations in medial knee OA. J. Orthop. Res. 27, 78–83. Brindle, R.A., Milner, C.E., Zhang, S., Fitzhugh, E.C., 2014. Changing step width alters lower extremity biomechanics during running. Gait Posture 39, 124–128.
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509
S. Kettlety et al. / Journal of Biomechanics xxx (xxxx) xxx Butler, R.J., Minick, K.I., Ferber, R., Underwood, F., 2009. Gait mechanics after ACL reconstruction: implications for the early onset of knee osteoarthritis. Br. J. Sports Med. 43, 366–370. Chang, A., Hayes, K., Dunlop, D., Song, J., Hurwitz, D., Cahue, S., Sharma, L., 2005. Hip abduction moment and protection against medial tibiofemoral osteoarthritis progression. Arthritis Rheum. 52, 3515–3519. Cortes, N., Quammen, D., Lucci, S., Greska, E., Onate, J., 2012. A functional agility short-term fatigue protocol changes lower extremity mechanics. J. Sports Sci. 30, 797–805. Eddo, O., Lindsey, B., Caswell, S.V., Cortes, N., 2017. Current evidence of gait modification with real-time biofeedback to alter kinetic, temporospatial, and function-related outcomes: a review. Int. J. Kinesiol. Sports Sci. 5, 35–55. Ewers, B.J., Jayaraman, V.M., Banglmaier, R.F., Haut, R.C., 2002. Rate of blunt impact loading affects changes in retropatellar cartilage and underlying bone in the rabbit patella. J. Biomech. 35, 747–755. Ferrigno, C., Stoller, I.S., Shakoor, N., Thorp, L.E., Wimmer, M.A., 2016. The feasibility of using augmented auditory feedback from a pressure detecting insole to reduce the knee adduction moment: a proof of concept study. J. Biomech. Eng. 138, 021014-1–021014-7. Gerbrands, T.A., Pisters, M.F., Theeven, P.J.R., Verschueren, S., Vanwanseele, B., 2017. Lateral trunk lean and medializing the knee as gait strategies for knee osteoarthritis. Gait Posture 51, 247–253. Grood, E.S., Suntay, W.J., 1983. A joint coordinate system for the clinical description of three-dimensional motions: application the knee. J. Biomech. Eng. 105, 136– 144. Hunt, M.A., Simic, M., Hinman, R.S., Bennell, K.L., Wrigley, T.V., 2011. Feasibility of a gait retraining strategy for reducing knee joint loading: increased trunk lean guided by real-time biofeedback. J. Biomech. 44, 943–947. Hunt, M.A., Wrigley, T.V., Hinman, R.S., Bennell, K.L., 2010. Individuals with severe knee osteoarthritis (OA) exhibit altered proximal walking mechanics compared with individuals with less severe OA and those without knee pain. Arthritis Care Res. 62, 1426–1432. Hunt, M., Birmingham, T., Bryant, D., Jones, I., Giffin, J., Jenkyn, T., Vandervoort, A., 2008. Lateral trunk lean explains variation in dynamic knee joint load in patients with medial compartment knee osteoarthritis. Osteoarth. Cartil. 16, 591–599. Kean, C.O., Bennell, K.L., Wrigley, T.V., Hinman, R.S., 2015. Relationship between hip abductor strength and external hip and knee adduction moments in medial knee osteoarthritis. Clin. Biomech. 30, 226–230. Miyazaki, T., Wada, M., Kawahara, H., Sato, M., Baba, H., Shimada, S., 2002. Dynamic load at baseline can predict radiographic disease progression in medial compartment knee osteoarthritis. Ann. Rheum. Dis. 61, 617–622. Mündermann, A., Asay, J.L., Mündermann, L., Andriacchi, T.P., 2008. Implications of increased medio-lateral trunk sway for ambulatory mechanics. J. Biomech. 41, 165–170.
7
Mündermann, A., Dyrby, C.O., Andriacchi, T.P., 2005. Secondary gait changes in patients with medial compartment knee osteoarthritis: increased load at the ankle, knee, and hip during walking. Arthritis Rheum. 52, 2835–2844. Paquette, M.R., Klipple, G., Zhang, S., 2015. Greater step widths reduce internal knee abduction moments in medial compartment knee osteoarthritis patients during stair ascent. J. Appl. Biomech. 31, 229–236. Reeves, N.D., Bowling, F.L., 2011. Conservative biomechanical strategies for knee osteoarthritis. Nat. Rev. Rheumatol. 7, 113. Richards, R.E., van den Noort, J.C., van der Esch, M., Booij, M.J., Harlaar, J., 2018. Effect of real-time biofeedback on peak knee adduction moment in patients with medial knee osteoarthritis: is direct feedback effective?. Clin. Biomech. 57, 150–158. Richards, R., van den Noort, J.C., Dekker, J., Harlaar, J., 2017. Gait retraining with real-time biofeedback to reduce knee adduction moment: systematic review of effects and methods used. Arch. Phys. Med. Rehabil. 98, 137–150. Rutherford, D.J., Hubley-Kozey, C., 2009. Explaining the hip adduction moment variability during gait: Implications for hip abductor strengthening. Clin. Biomech. 24, 267–273. Schwartz, M.H., Rozumalski, A., 2005. A new method for estimating joint parameters from motion data. J. Biomech. 38, 107–116. Sharma, L., Hurwitz, D.E., Thonar, E.J.-M.A., Sum, J.A., Lenz, M.E., Dunlop, D.D., Schnitzer, T.J., Kirwan-Mellis, G., Andriacchi, T.P., 1998. Knee adduction moment, serum hyaluronan level, and disease severity in medial tibiofemoral osteoarthritis. Arthritis Rheum. 41, 1233–1240. Shull, P.B., Shultz, R., Silder, A., Dragoo, J.L., Besier, T.F., Cutkosky, M.R., Delp, S.L., 2013a. Toe-in gait reduces the first peak knee adduction moment in patients with medial compartment knee osteoarthritis. J. Biomech. 46, 122–128. Shull, P.B., Silder, A., Shultz, R., Dragoo, J.L., Besier, T.F., Delp, S.L., Cutkosky, M.R., 2013b. Six-week gait retraining program reduces knee adduction moment, reduces pain, and improves function for individuals with medial compartment knee osteoarthritis. J. Orthop. Res. 31, 1020–1025. Simic, M., Hinman, R.S., Wrigley, T.V., Bennell, K.L., Hunt, M.A., 2011. Gait modification strategies for altering medial knee joint load: a systematic review. Arthritis Care Res. 63, 405–426. Simic, M., Hunt, M.A., Bennell, K.L., Hinman, R.S., Wrigley, T.V., 2012. Trunk lean gait modification and knee joint load in people with medial knee osteoarthritis: the effect of varying trunk lean angles. Arthritis Care Res. 64, 1545–1553. Tateuchi, H., Koyama, Y., Akiyama, H., Goto, K., So, K., Kuroda, Y., Ichihashi, N., 2017. Daily cumulative hip moment is associated with radiographic progression of secondary hip osteoarthritis. Osteoarth. Cartil. 25, 1291–1298. Vigotsky, Andrew D., Zelik, K.E., Lake, J., Hinrichs, R.N., 2019. Mechanical misconceptions: have we lost the ‘‘mechanics” in ‘‘sports biomechanics”?. J. Biomech. 93, 1–5. Winter, D., 2009. Biomechanics and Motor Control of Human Movement. John Wiley & Sons Inc, Hoboken.
Please cite this article as: S. Kettlety, B. Lindsey, O. Eddo et al., Changes in hip mechanics during gait modification to reduce knee abduction moment, Journal of Biomechanics, https://doi.org/10.1016/j.jbiomech.2019.109509