0
Research Papers

Predicting Sagittal Plane Biomechanics That Minimize the Axial Knee Joint Contact Force During Walking

[+] Author and Article Information
Ross H. Miller

e-mail: rosshm@umd.edu

Kevin J. Deluzio

Department of Mechanical and
Materials Engineering,
Queen's University,
Kingston, ON, K7L 3N6, Canada

1Corresponding author. Present address: Department of Kinesiology, University of Maryland, 2134 A SPH Building, College Park, MD 20742.

Contributed by the Bioengineering Division of ASME for publication in the Journal of Biomechanical Engineering. Manuscript received May 29, 2012; final manuscript received December 8, 2012; accepted manuscript posted December 15, 2012; published online December 27, 2012. Assoc. Editor: Mohamed Samir Hefzy.

J Biomech Eng 135(1), 011007 (Dec 27, 2012) (11 pages) Paper No: BIO-12-1211; doi: 10.1115/1.4023151 History: Received May 29, 2012; Revised December 08, 2012; Accepted December 15, 2012

Both development and progression of knee osteoarthritis have been associated with the loading of the knee joint during walking. We are, therefore, interested in developing strategies for changing walking biomechanics to offload the knee joint without resorting to surgery. In this study, simulations of human walking were performed using a 2D bipedal forward dynamics model. A simulation generated by minimizing the metabolic cost of transport (CoT) resembled data measured from normal human walking. Three simulations targeted at minimizing the peak axial knee joint contact force instead of the CoT reduced the peak force by 12–25% and increased the CoT by 11–14%. The strategies used by the simulations were (1) reduction in gastrocnemius muscle force, (2) avoidance of knee flexion during stance, and (3) reduced stride length. Reduced gastrocnemius force resulted from a combination of changes in activation and changes in the gastrocnemius contractile component kinematics. The simulations that reduced the peak contact force avoided flexing the knee during stance when knee motion was unrestricted and adopted a shorter stride length when the simulated knee motion was penalized if it deviated from the measured human knee motion. A higher metabolic cost in an offloading gait would be detrimental for covering a long distance without fatigue but beneficial for exercise and weight loss. The predicted changes in the peak axial knee joint contact force from the simulations were consistent with estimates of the joint contact force in a human subject who emulated the predicted kinematics. The results demonstrate the potential of using muscle-actuated forward dynamics simulations to predict novel joint offloading interventions.

FIGURES IN THIS ARTICLE
<>
Copyright © 2013 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Diagram of the musculoskeletal model showing the rigid body segments, right leg muscle models, and ground contact elements

Grahic Jump Location
Fig. 2

Joint angles (hip, knee, ankle) and GRF (horizontal, vertical) for the tracking simulation (thick lines) and for the human subjects (thin lines). No experimental data were available for the toe joint angle. The stride (x-axis) begins and ends at heel-strike.

Grahic Jump Location
Fig. 3

Portions of the stride when the muscle models were on (nonzero excitation; black bars) for the data tracking simulation (“Track”). “EMG” are nominal timing data derived from electromyograms by Knutson and Soderberg [27]. The stride (x-axis) begins and ends at heel-strike.

Grahic Jump Location
Fig. 4

Stick figures of walking motions for the four predictive simulations, plus descriptive metrics. (a) Minimized metabolic cost, (b) minimized peak axial knee joint contact force, (c) minimized peak knee flexion angle in stance, (d) minimized peak axial knee joint contact force and tracked the mean experimental knee angle. Speed = average horizontal speed, SL = stride length, SF = stride frequency, CoT = metabolic cost of transport, and SD = average deviation from the experimental joint angles and GRF.

Grahic Jump Location
Fig. 5

Joint angles (in deg) and GRF (in multiples of bodyweight) for each of the four predictive simulations. (a) Minimized metabolic cost, (b) minimized peak axial knee joint contact force, (c) minimized peak knee flexion angle in stance, (d) minimized peak axial knee joint contact force and tracked the mean experimental knee angle. No experimental data were available for the toe joint angle. The stride (x-axis) begins and ends at heel-strike.

Grahic Jump Location
Fig. 6

Muscle model activations for each of the four predictive simulations. (a) Minimized metabolic cost, (b) minimized peak axial knee joint contact force, (c) minimized peak knee flexion angle in stance, (d) minimized peak axial knee joint contact force and tracked the mean experimental knee angle. ILP = iliopsoas, GLU = glutei, VAS = vasti, BFS = biceps femoris (short head), TA = tibialis anterior, SOL = soleus, RF = rectus femoris, HAM = hamstrings, and GAS = gastrocnemius. The stride (x-axis) begins and ends at heel-strike.

Grahic Jump Location
Fig. 7

Axial knee joint contact forces (thick line) the four predictive simulations. (a) Minimized metabolic cost, (b) minimized peak axial knee joint contact force, (c) minimized peak knee flexion angle in stance, (d) minimized peak axial knee joint contact force and tracked the mean experimental knee angle. Thin solid line (“Knee mus/lig”) is the contribution from muscles and ligaments spanning the knee. Thin broken line (“Reaction”) is the contribution from the knee joint reaction force (i.e., the resultant joint force from inverse dynamics). The stride (x-axis) begins and ends at heel-strike. The directions of the joint reaction and contact forces are flipped for presentation.

Grahic Jump Location
Fig. 8

Knee muscle force components along the long axis of the tibia from the four predictive simulations. Column (a) is minimized metabolic cost; Column (b) is minimized peak axial knee joint contact force; Column (c) is minimized peak knee flexion angle in stance; Column (d) is minimized peak axial knee joint contact force and tracked the mean experimental knee angle. VAS = vasti, BFS = biceps femoris (short head), RF = rectus femoris, HAM = hamstrings, and GAS = gastrocnemius. The stride (x-axis) begins and ends at heel-strike.

Grahic Jump Location
Fig. 9

Changes in axial knee muscle impulses (force-time integral) between the minimum contact force simulations and the minimum metabolic cost simulation. White bars are changes due to muscle activation level. Black bars are changes due to muscle kinematics. (a) Simulation that minimized peak axial knee joint contact force, (b) simulation that minimized peak knee flexion angle in stance, (c) simulation that minimized peak axial knee joint contact force and tracked the mean experimental knee angle.

Grahic Jump Location
Fig. 10

Peak axial knee joint contact forces from three of the predictive simulations (black bars) and estimated using static optimization for a single human subject (white bars) who walked while attempting to emulate the kinematics shown in Figs. 4(c) and 4(d) (“Straight Knee” and “Quicker Steps”). Error bars are ± one standard deviation between ten trials.

Grahic Jump Location
Fig. 11

Axial knee joint contact forces from two additional simulations that used a nodal control scheme (e.g., Miller et al. [21]) with 41 parameters per muscle. The solid line minimized the sum of the joint angle and GRF tracking errors and the metabolic cost. The dashed line minimized the peak of the contact force. The stride (x-axis) begins and ends at heel-strike.

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In