Determinants of Human Gait: A Review

1
Determinants of Human Gait A Review Role of
Knee/Ankle Coupling in Stability, Control
Propulsion Gordon J. Alderink, PT, PhD Grand
Valley State University Cook-DeVos Center for
Health Sciences Grand Rapids, Michigan USA
Background Saunders and co-workers originally
described six determinants (D1, D2, D3, etc) of
gait as precise movements by the pelvis, hip,
knee and ankles that theoretically minimized
vertical and horizontal excursion of the bodys
center of mass (COM), thus, reducing the energy
cost of walking. However, it has recently been
suggested that although these movements certainly
occur, some of them may play little or no part in
optimizing energy cost. Furthermore, there is
evidence that a flattened COM trajectory
increases muscle work and force requirements.
Proponents of the dynamic gait perspective
suggest that an inverted pendulum model of gait
better explains the mechanical work and,
therefore, metabolic costs of walking. Because of
the complexity of human gait, mathematical models
to describe or simulate normal walking have been
justifiably simplified. For example, Saunderss
determinants of gait provide only a
one-dimensional explanation of how humans may
control energy expenditure while walking. Since
recent evidence suggests that these gait
determinants may not play a major role in
controlling energy cost, one might examine gait
determinants from a different perspective.
Relevant to humanoid research energy cost may not
be as important to consider as propulsion and
control (stability). Furthermore, we might need
to consider the interdependence of kinematic,
kinetic and dynamic factors with regard to energy
cost and control. Purpose The purposes of this
presentation are 1) review the determinants of
gait 2) review the dynamic walking perspective
(inverted pendulum model) 3) review static and
dynamic postural/gait controls and 4) consider
knee/ankle coupling (D3, D4 D5) as crucial
determinants for a stable, smooth dynamic human
gait. Determinants of Gait D1. Pelvic Rotation.
Rotation of the pelvis about a vertical axis
reduces the angle of hip flexion and extension,
minimizing the rise and fall of the hip joint,
and, thus, elevation of COM during a stride. D2.
Pelvic Obliquity. If the pelvis were to remain
level during a stride, the rise and fall of the
hip joint associated with flexion and extension
would force the trunk to rise and fall as a
function of the average elevation of both hips
(stance and swing). Pelvic tipping about an
antero-posterior axis resulting in a downward
slope of the pelvis toward the swing leg reduces
the cranial excursion of the trunk. D3. Knee
Flexion in Stance. Early stance knee flexion
effectively keeps hip height constant, thus
reducing the height of the apex of the COM. D4.
Ankle Mechanism. The apex of the COM is
lengthened at initial contact by a dorsiflexed
ankle (1st rocker). D5. Foot Mechanism. The leg
is lengthened at the end of stance as the ankle
moves from dorsiflexion to plantarflexion (heel
rise or 3rd rocker), thus reducing COM vertical
displacement. D6. Lateral Displacement of Body.
Slight physiologic knee valgus reduces the
walking base of support (BOS), thus minimizing
side to side displacement of the COM. Kuo
suggests that a flattened COM, as dictated by the
six determinants, increases the muscle work,
force requirements, and, therefore, energy costs
of walking. Although the determinants do reduce
COM excursion in a compass gait, Della Croce et
al., Kerrigan et al., and Gard Childress
concluded that most determinants play little or
no role in reducing COM and energy cost.
Kerrigan et al. demonstrated that only D5
optimized the height of COM. Baker et al.
suggested the optimization of energy expenditure
during gait was not related to lowering the COM,
but related to maintaining phase relationship and
relative amplitude of the gravitational and
kinetic energy of the body

• Dynamic Gait Perspective
• Dynamic walking simplifies the study of gait and
  offers a constructive perspective, i.e., yields
  predictions independent of experimental data.
  Since the determinant model fares poorly, Kuo
  suggests examining an inverted pendulum model
• The single support leg behaves like an inverted
  pendulum to transport the COM with relatively
  little muscle force and work (much less than the
  gait favored by the determinants theory)
• Walking like an inverted pendulum requires a
  step-to-step transition, which require work to
  redirect COM velocity
• Forced leg motion produces a trade-off in
  step-to-step transition costs vs energy cost
  related to force production
• Kuo proposed a refined interpretation of the
  inverted pendulum gait using muscular-driven
  models that can be described using four intervals
  of stance phase (Figure 1).

• Role of D3, D4 D5 in Stability, Control
  Propulsion
• The knee and ankle/foot are comprised of 30
  synovial joints with 6 DOF movement. Each joint
  plays a unique interdependent role in the
  initiation and maintenance of a stable,
  controlled, smooth efficient gait.
• Movement is produced/controlled actively (muscle)
  and passively (joint morphology periarticular
  soft tissues). Muscle stiffness is controlled by
  its material and active intrinsic properties, and
  reflexes (joint mechanoreceptors, GTOs and
  muscle spindles). Muscle actions account for 50
  to 95 of the vertical ground reaction force
  (GRF) generated in stance phase GRFs translate
  into relatively high joint reactions forces,
  e.g., 2.5 x BW at hip in single limb support.
  Physical Stress Theory suggests that the human
  body will attempt to attenuate high joint
  stresses.
• Static (posture) and dynamic (gait) balance is
  provided by ankle/hip and hip/knee/ankle
  strategies, as well as visual and vestibular
  input. During gait reflex activity (at a
  metabolic cost) at the hip, knee and ankle
  control antero-posterior, and at the hip control
  medio-lateral, acceleration of the head, arms and
  trunk, at the same time other essential kinematic
  events are taking place, e.g., joint motion, step
  length, toe clearance, etc.
• Lets also examine D3, D4 D5 and muscle
  requirements, using the refined inverted pendulum
  model proposed by Kuo (Figure 1)
• From collision to rebound (initial contact
  through loading response), the knee is flexing as
  the ankle is plantarflexing. During this
  subphase hip and knee extensors are main
  contributors early in stance, as are the ankle
  dorsiflexors.
• From rebound through preload (midstance to
  terminal stance) the knee remains extended as the
  tibia moves over the fixed foot (ankle
  dorsiflexion). The gluteus maximus, vasti,
  soleus and posterior gluteus medius make
  substantial contributions to knee extension,
  while the ankle plantarflexors provide primary
  support in late stance and is a major factor in
  producing forward body progression.
• From pre-load through push-off (terminal stance
  to preswing) the knee rapidly flexes as the ankle
  begins to plantarflex. During this time period,
  the iliopsoas and gastrocnemius are the largest
  contributors to peak knee flexion velocity during
  double support. Apparently, the sartorius and
  gracilis can assist in producing optimal knee
  angular velocity.
• In conclusion, it appears likely that D3, D4 D5
  are important determinants to control COM
  excursion, metabolic costs, joint stresses, and
  provide stability. Robotic (humanoid) research
  might be furthered as a profound understanding of
  the interdependent nature of human gait mechanics
  is realized.
• References
• Baker R et al., 8th International Symposium on
  the 3-D Analysis of Human Movement, 2004.
• Della Croce U et al., Gait Posture, 14 79-84,
  2001.
• Ferber R et al., Gait Posture, 16 238-248,
  2002.
• Gard S and Childress D, Gait Posture, 5
  233-238, 1997.
• Gard S and Childress D, Arch Phys Med Rehabil,
  80 26-32, 1999.
• Kerrigan C et al., Arch Phys Med Rehabil, 82
  217-220, 2001.
• Kuo A et al., Exerc Sport Sci Rev, 33 88-97,
  2005.
• Kuo A, Human Movement Science, 26 617-656, 2007.
• Magee D et al., Scientific Foundations and
  Principles of Practice in Musculoskeletal
  Rehabilitation, Saunders Elsevier, 2007.

Figure 1. Four subphases of stance illustrating
instances of joint work and trajectory of COM
(Kuo A et al., Exerc. Sport Sci. Rev. 33 (2),
88-97, 2005).

• Work is required to redirect the COM between
  pendular arcs so that positive work is performed
  by the trailing leg before or simultaneous with
  negative work by the leading leg. Metabolic
  cost depends not on COM displacement per se, but
  on COM redirection between steps and the rate of
  work and metabolic energy expenditure are
  related to step length and width.
• With the inverted pendulum model sagittal plane
  passive dynamic properties may provide
  stability. However, when more degrees of freedom
  are added to the model significant active
  control may be needed to stabilize lateral
  motion.
• It can be argued that those utilizing a dynamic
  walking model (a compass gait in itself)
  misinterpreted Saunders et al. explanation for a
  relatively flat COM trajectory. Dynamic
  walking replaces one simple model with another
  one, which certainly can produce complete gaits,
  but cannot model human gait complexity.
  Muscle-driven forward simulations of normal and
  pathological gait call into question the ability
  of simple dynamic models to characterize gait.
  For example, muscle models incorporating
  force-length and force- velocity properties of
  muscle can best explain static and dynamic biped
  perturbations. Furthermore, dynamic simulations
  to perform muscle- induced segmental acceleration
  and power analyses have shown
• 1) muscles do substantial work in raising the
  COM in early stance, and 2) the interdependency
  of joint power transfers. Finally, one-, two-
  and three- dimensional dynamic models, because of
  their simplicity, do not account for the
  interdependent role of joint receptors, soft
  tissue controls (ligament and muscle), and 6 DOF
  joint movements.
• While I concur that simple models can be
  constructive, they do not take into account the
  multitasking nature of the integrated
  neuro-sensory- musculo-skeletal human that
  locomotes smoothly, while minimizing physical
  stress, i.e., Physical Stress Theory, and
  metabolic costs. Therefore, lets examine, in a
  different way, three of the gait determinants.