2362_Ch14-524-568.

qxd 2/2/11 4:44 PM Page 524

Chapter
14
Gait
Sandra J. Olney, PT, OT, PhD, and Janice Eng, PT, OT, PhD

Introduction Trunk and Upper Extremities

Gait Analysis Trunk
Major Tasks of Gait Upper Extremities
Phases of the Gait Cycle Treadmill, Stair, and Running Gaits
Events in Stance Phase
Treadmill Gait
Subphases of Stance Phase
Stair Gait
Swing Phase
Running Gait
Gait Terminology Joint Motion and Muscle Activity
Time and Distance Terms Moments, Powers, and Energies
Kinematic Terms Summary
Kinetic Terms Effects of Age, Gender, Assistive Devices,
Electromyography and Orthoses
Characteristics of Normal Gait Age
Time and Distance Characteristics Gender
Sagittal Plane Joint Angles Assistive Devices
Frontal Plane Joint Angles Orthoses
Ground Reaction Force and Center of Pressure Abnormal Gait
Sagittal Plane Moments Structural Impairment
Frontal Plane Moments Functional Impairment
Sagittal Plane Powers Pain
Frontal Plane Powers Adaptation/Compensation
Mechanical Energy of Walking
Muscle Activity
Gait Initiation and Termination

524
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 525

CHAPTER 14 ■ Gait 525

INTRODUCTION

Gait Analysis
In human locomotion (ambulation, gait), the reader is
given the opportunity to discover how individual joints
and muscles function in an integrated manner both to
maintain upright posture and to produce motion of the
body as a whole. Knowledge of the kinematics and kinet-
ics of normal ambulation provides the reader with a
foundation for analyzing, identifying, and correcting
abnormalities in gait.
Walking is probably the most comprehensively studied
of all human movements, and the variety of technologies,
coupled with the diversity of disciplinary perspectives, has
produced a complex and sometimes daunting literature.
The biomechanical requirements of the movements that
explain gait are logical and easily understood if the detail is
not permitted to cloud comprehension. The purpose of this Figure 14-1 A modern gait laboratory.
chapter is to provide this comprehension of gait that will
serve as the foundation for analysis of normal walking and
of gait deviations. remaining 50%.4 To make full use of this chapter you
In early gait analysis, investigators used cinematographic should review the relevant biomechanics and anatomy.
film and until about 20 years ago, sophisticated analysis re- You must understand the basic biomechanical concepts
quired frame-by-frame hand-digitizing of markers that had presented in Chapter 1, the joint motion of major joints of
been placed on body landmarks. These data were coupled the lower limbs, and have a thorough knowledge of the
with knowledge of the center of pressure (CoP) of the foot- major muscle groups of the lower limbs and their actions.
floor forces derived from a force platform to give complete,
if simplified, kinetic information. This is referred to as the Major Tasks of Gait
inverse dynamic approach with link segment mechanics. To understand gait, let us first identify the fundamental pur-
Electrogoniometers fastened to joints were also commonly poses. Winter5 proposed the following five main tasks for
used to describe joint motion and still have applications.1 walking gait:
Similarly, electromyography (EMG) has been used for many
decades, although the expectation that it would be possible 1. Maintenance of support of the head, arms, and trunk,
to convert those signals to force values in simple, useful that is, preventing collapse of the lower limb
ways has not been fulfilled. However, the past two decades 2. Maintenance of upright posture and balance of the body
have witnessed an explosion of technical advancements in 3. Control of the foot trajectory to achieve safe ground
motion analysis that offer the ability to collect and process clearance and a gentle heel or toe landing
large amounts of data. As with the development of any 4. Generation of mechanical energy to maintain the
science, the knowledge available far exceeds its current present forward velocity or to increase the forward
applications. velocity
A modern gait laboratory (Fig. 14–1) includes some kind 5. Absorption of mechanical energy for shock absorption
of motion analysis system that employs precise marker loca- and stability or to decrease the forward velocity of the
tions that are subsequently used to model a several-segment body
body with joint centers and centers of mass. Also included Maintaining balance and stability is clearly important
are one or more force platforms that provide simultaneous during ambulation, and there is an increasing body of
foot-floor forces. EMG systems provide simultaneous infor- literature on what might be called sub-tasks of gait that
mation from surface electrodes, or, sometimes, indwelling are potentially destabilizing. These include gait initiation,6
electrodes. An excellent and engaging report of the evolu- and termination,7 stair-climbing,8 turning,9 obstacle cross-
tion of clinical gait analysis, including motion analysis and ing,9–11 and negotiating a raised surface.12 Gait initiation
EMG, can be found in Sutherland’s articles.2,3 and termination and stair-climbing will be introduced in
Human locomotion, or gait, may be described as a this chapter.
translatory progression of the body as a whole, produced
by coordinated, rotatory movements of body segments.1
The alternating movements of the lower extremities essen-
Phases of the Gait Cycle
tially support and carry along the head, arms, and trunk. Gait has been divided into a number of segments that make
The head, arms, and trunk constitute about 75% of total it possible to describe, understand, and analyze the events
body weight, with the head and arms contributing about that are occurring. A gait cycle spans two successive events of
25% of total body weight and the trunk contributing the the same limb, usually initial contact of the lower extremity
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 526

526 SECTION 5 ■ Integrated Function

with the supporting surface. During one gait cycle, each ex- events and phases. Values for normal walking appear in
tremity passes through two major phases: a stance phase, the figures.
when some part of the foot is in contact with the floor, which
makes up about 60% of the gait cycle,13 and a swing phase,
when the foot is not in contact with the floor, which makes up
Events in Stance Phase
the remaining 40%13,14 (Fig. 14–2). There are two periods of 1. Initial contact refers to the instant the foot of the leading
double support occurring between the time one limb makes extremity strikes the ground.16 In normal gait, the heel
initial contact and the other one leaves the floor at toe-off. is the point of contact, and the event referred to as heel
At a normal walking speed, each period of double support contact or heel strike. The word strike is actually a mis-
occupies about 11% of the gait cycle, which makes a total of nomer inasmuch as the horizontal velocity reduces
approximately 22% for a full cycle.15 The body is thus sup- to about 0.4 m/sec and only 0.05 m/sec vertically.17 In
ported by only one limb for nearly 80% of the cycle. The abnormal gait, it is possible for the whole foot or the
approximate value of 10% for each double-support phase is toes, rather than the heel, to make initial contact with
usually assigned to each of the two double-support periods. the ground.
Stance phase is divided into subphases by a number 2. Foot flat in normal gait occurs after initial contact at
of events that mark the start and end of the subphases. approximately 7% of the gait cycle. It is the first instant
Figure 14–3 identifies the events delimiting the stance during stance when the foot is flat on the ground.
phase as initial contact, sometimes referred to as heel 3. Midstance is the point at which the body weight is
contact or heel strike and toe-off. The gait cycle is directly over the supporting lower extremity, usually
divided into percentiles that will be used to clarify about 30% of the gait cycle.

Right Right Right Left Right Right Right Figure 14-2 A gait cycle spans
Right
initial foot flat 7% midstance heel off initial toe off midswing initial the period between initial contact
contact 30% 40% contact 60% contact of the reference extremity (right)
0% Left toe off 50% 100%
10% and the successive contact of the
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
same extremity. This figure
shows the gait cycle with major
events: stance and swing phases
GAIT CYCLE for each limb and periods of
single and double support. The
stance phase constitutes 60% of
RIGHT STANCE RIGHT SWING
the gait cycle, and the swing
Right phase constitutes 40% of the
stance LEFT SWING LEFT STANCE cycle at normal walking speeds.
Increases or decreases in walking
Double Right single Double Left single speeds alter the percentages of
support limb support support limb support time spent in each phase.

Figure 14-3 The stance

phase of a gait cycle of the
% Gait right lower limb. The events
Cycle 0% 10% 20% 30% 40% 50% 60% that delimit subphases are
shown and expressed as
Initial Foot
Event

Midstance Heel off Toe off percentages of full gait cycle:

contact flat 7% 30% 40% 60%
0%
initial contact, foot flat,
heel-off, midstance, and
toe-off. Subphases are heel
Phase

Heel strike Midstance Push off strike phase, midstance

phase, and push-off phase.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 527

CHAPTER 14 ■ Gait 527

4. Heel-off is the point at which the heel of the reference

extremity leaves the ground, usually about 40% of the
gait cycle.
5. Toe-off is the instant at which the toe of the foot leaves
the ground, usually about 60% of the gait cycle.

Subphases of Stance Phase

1. Heel strike phase begins with initial contact and ends with
foot flat and occupies only a small percentage of the gait
cycle (see Fig. 14–3). Terminology originating in Rancho Early swing Midswing Late swing
Los Amigos National Rehabilitation Center includes this 60-75% 75-85% 85-100%
phase in one called weight acceptance, or loading response, which
begins at initial contact and ends when the contralateral Figure 14-4 The swing phase of a gait cycle. Early swing phase
is also known as initial swing or acceleration, and late swing is
extremity lifts off the ground at the end of the double-
also known as terminal swing or deceleration.
support phase16 and occupies about 11% of the gait cycle.
2. Midstance phase begins with foot flat at 7% of the gait cycle
and ends with heel-off at about 40% of the gait cycle.
3. Push-off phase begins with heel-off at about 40% of the GAIT TERMINOLOGY
gait cycle and ends with toe-off at about 60% of the gait
cycle (see Fig. 14–3). Time and Distance Terms
Sometimes authors refer to a preswing phase, which is Time and distance are two basic parameters of motion, and
terminology originating in Rancho Los Amigos National measurements of these variables provide a basic description
Rehabilitation Center.16 Preswing refers to the last 10% of of gait. Temporal variables include stance time, single-limb
stance phase, beginning with initial contact of the contralat- and double-support time, swing time, stride and step time,
eral foot (at 50% of the gait cycle) and ending with toe-off cadence, and speed. The distance variables include stride
(at 60%). length, step length and width, and degree of toe-out. These
variables, derived in classic research of over 30 years ago,
Swing Phase provide essential quantitative information about a person’s
gait and should be included in any gait description.16,18–21
1. Early swing phase begins once the toe leaves the ground
Each variable may be affected by such factors as age, sex,
and continues until midswing, or the point at which
height, size and shape of bony components, distribution of
the swinging extremity is directly under the body
mass in body segments, joint mobility, muscle strength, type
(Fig. 14–4). This phase is also referred to as initial swing,
of clothing and footgear, habit, and psychological status.
or the acceleration phase.
However, a discussion of all the factors affecting gait is be-
2. Midswing occurs approximately when the extremity
yond the scope of this text.
passes directly beneath the body, or from the end of
Stance time is the amount of time that elapses during
acceleration to the beginning of deceleration.
the stance phase of one extremity in a gait cycle.
3. Late swing occurs after midswing when the limb is decel-
Single-support time is the amount of time that elapses
erating in preparation for heel strike. It is also known as
during the period when only one extremity is on the sup-
terminal swing, or the deceleration phase.
porting surface in a gait cycle.
Double-support time is the amount of time spent with
14-1 Patient Case
case
Marlene Brown is a 63-year-old woman who sustained a
both feet on the ground during one gait cycle. The percent-
age of time spent in double support may be increased in
elderly persons and in those with balance disorders. The
stroke 15 days ago and shows weakness, or hemiparesis, of percentage of time spent in double support decreases as the
her right arm and leg. She has been in a rehabilitation unit speed of walking increases.
for 10 days and is making good progress walking, although Stride length is the linear distance between two succes-
she is ambulating at only 0.20 m/sec. A value of about sive events that are accomplished by the same lower extrem-
1 m/sec would be typical for an able-bodied person of her ity during gait.13 In general, stride length is determined by
age. She has weakness in several muscle groups of her lower measuring the linear distance from the point of one heel
limb, notably the ankle plantarflexors, ankle dorsiflexors, strike of one lower extremity to the point of the next heel
knee extensors, and hip flexors, with distal muscles more strike of the same extremity (Fig. 14–5). The length of one
affected than proximal ones. Particularly troublesome is the stride includes all of the events of one gait cycle. Stride
inability to clear her foot during the swing phase of gait. She length also may be measured by using other events of the
uses a cane with a large base (a four-point cane) in her same extremity, such as toe-off, but in normal gait, two suc-
unaffected hand for stability. As you read the Gait Terminology cessive heel strikes are usually used. A stride includes two
section, ask youself what differences from normal gait you steps, a right step and a left step. However, stride length is
might expect to see. not always twice the length of a single step, because right and
left steps may be unequal. Stride length varies greatly among
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 528

528 SECTION 5 ■ Integrated Function

Right step length Left step length

Left foot angle

Figure 14-5 Stride length, step length,

Right foot angle and width shown with foot angle place-
ments. The midpoint of the heel is used
as a point of reference for measuring
Stride length step width.

individuals, because it is affected by leg length, height, age, inasmuch as the velocities of the segments involve specifi-
sex, and other variables. Stride length can be normalized cation of direction:
by dividing stride length by leg length or by total body
Walking velocity (meters/second) = distance walked
height, so people of different sizes can be compared. Stride
(meters)/time (seconds)
length usually decreases in elderly persons14,15,22 and in-
creases as the speed of gait increases.23 Women tend to walk with shorter and faster steps than
Stride duration refers to the amount of time it takes to do men at the same velocity23 due largely to height and leg
accomplish one stride. Stride duration and gait cycle dura- length differences. Increases in velocity up to 120 steps per
tion are synonymous. One stride, for a normal adult, lasts minute are brought about by increases in both cadence and
approximately 1 second.24 stride length, but above 120 steps per minute, step length
Step length is the linear distance between two successive levels off, and speed increases are achieved with only ca-
points of contact of opposite extremities. It is usually meas- dence increases.
ured from the heel strike of one extremity to the heel strike A person’s normal comfortable speed of gait may be
of the opposite extremity (see Fig. 14–5). A comparison of referred to as preferred, natural, self-selected, or free.
right and left step lengths will provide an indication of gait Slow and fast speeds of gait refer to speeds slower or faster
symmetry. The more equal the step lengths, the more sym- than the person’s normal comfortable walking speed,
metrical is the gait. designated in a variety of ways. There is a certain amount
Step duration refers to the amount of time spent of variability in the way an individual elects to increase
during a single step. Measurement usually is expressed as walking speed. Some individuals increase stride length to
seconds per step. When there is weakness or pain in an ex- achieve a fast walking speed. Others increase cadence.
tremity, step duration may be decreased on the affected Step width, or width of the walking base, may be found
side and increased on the unaffected (stronger) or less by measuring the linear distance between the midpoint of
painful side. the heel of one foot and the same point on the other foot
Cadence is the number of steps taken by a person per (see Fig. 14–5). Step width has been found to increase when
unit of time. Cadence may be measured as the number of there is an increased demand for side-to-side stability, such
steps per second or per minute, but the latter is more com- as occurs in elderly persons and in small children. In tod-
mon. A shorter step length will result in an increased dlers and young children, the center of gravity is higher
cadence at any given velocity.23 Lamoreaux found that when than in adults, and a wide base of support is necessary for
a person walks with a cadence between 80 and 120 steps per stability. In the normal population, the mean width of the
minute, cadence and stride length had a linear relation- base of support is about 3.5 inches and varies within a range
ship.13 As a person walks with increased cadence, the dura- of 1 to 5 inches.
tion of the double-support period decreases. When the Degree of toe-out represents the angle of foot place-
cadence of walking approaches 180 steps per minute, the ment and may be found by measuring the angle formed by
period of double support disappears, and running com- each foot’s line of progression and a line intersecting the
mences. A step frequency or cadence of about 110 steps per center of the heel and the second toe. The angle for men
minute can be considered as “typical” for adult men; a typi- normally is about 7° from the line of progression of each
cal cadence for women is about 116 steps per minute.4 foot at free speed walking (see Fig. 14–5).14 The degree of
Sometimes authors report values that refer to stride cadence, toe-out decreases as the speed of walking increases in nor-
which is exactly half the step cadence. mal men.14
Walking velocity is the rate of linear forward motion
of the body, which can be measured in meters or centime- Kinematic Terms
ters per second, meters per minute, or miles per hour.
Scientific literature favors meters per second. The term Kinematics is the term used to describe movements without
velocity implies that direction is specified, although this is considering the internal or external forces that caused the
frequently not included, and the more correct term walk- movements. These measures include positions, velocities,
ing speed should be used if direction is not reported. and accelerations of body markers or body segments. Sophis-
In instrumented gait analyses, walking velocity is used, ticated equipment—at first, stroboscopic photography, then
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 529

CHAPTER 14 ■ Gait 529

cinematography and electrogoniometers, and, more recently,

many types of computerized motion analysis systems—have
provided comprehensive information about limb positions
and their motions in normal and abnormal gait.2,3,25–27
Observational gait analysis employs a less sophisticated 1/2 Head, arms, and trunk
and less objective method in which an observer makes a segments:
mass
judgment as to whether a particular joint angle or motion
varies from a norm. Usually observational gait analysis is
used to hypothesize causes of deviations and to direct treat-
Moment of inertia
ment objectives. One disadvantage of the observational location of center of mass
method of analysis is that it requires a great deal of training
and practice to be able to identify the particular segment of
Thigh segment:
gait in which a particular joint angle deviates from a norm mass
while a person is walking. Videotaping the action and then moment of inertia
viewing it using slow playback can improve this greatly. An- location of center of mass
other disadvantage of observational gait analysis methods is
that they frequently have low reliability, although recent re- Lower leg segment:
ports have identified some variables and conditions under mass
moment of inertia
which reliability is satisfactory.28 location of center of mass
Trajectories, or the paths in space of particular parts, are
of particular interest. For example, the path of the toe dur-
ing gait is important in research into trips and falls. Foot segment:
Joint angles are of primary importance in recording gait. mass
They may be expressed as absolute positions in space or, more moment of inertia
location of center of mass
commonly, as the relative angles of joints between adjacent
segments, such as the hip or the knee. In most literature the Figure 14-6 Simple four-segment link segment model of half a
anatomical position is called the “zero” position of the joint. human body for use in gait studies. For each segment identified,
the mass, moment of inertia, and location of the center of mass
Kinetic Terms should be known for applications.

Kinetics is concerned with the forces acting on the body

that are the cause of the movement. A kinetic analysis is per-
formed to understand the forces acting on the foot by the body segment by segment and solve for the more proximal
supporting surface, the forces acting on the joints, the forces joint. Larger numbers of segments and three-dimensional
produced by muscles, the moments produced by those mus- analyses are more complex than this, but the principles are
cles crossing the joints, the mechanical power generated or the same.
absorbed by those muscles, and energy patterns of the body Ground reaction forces (GRFs) are the forces being ap-
during walking. plied to the foot by the ground when a person takes a step.
A link-segment model is most often used with an in- These forces are equal in size (magnitude) but opposite in
verse dynamic approach. This means that we look at the direction to the forces applied to the ground by the foot.
body as a series of segments (e.g., foot, lower leg, thigh, etc.) GRFs are expressed using vertical, anteroposterior, and
with links at the joints. We begin the analysis starting with mediolateral axes.4,31 If we combine the force components in
the segment touching the supporting surface (foot) and use two or thee planes, the vector sum is a single expression of
Newtonian mechanics (Fig. 14–6).To do this we must know the ground reaction force and is termed the ground reaction
the positions of body markers through the gait cycle, usually force vector (GRFV). When we refer to a single component,
from the motion analysis technology. We also need to know such as the vertical GRF, the abbreviation GRF will be used.
the forces acting on the body, and for the walking person, If components are combined into a single vector (force[s]
these are the ground reaction forces, derived from a force with a direction), GRFV will be used. In gait we frequently
platform,36 the weight of the body parts (gravity), and iner- use the sagittal GRFV. The center of pressure of the foot
tia of the body parts.4,30 Inertia (units kg ⫻ m2) is a measure on the supporting surface is the point where the resultant of
of an object’s resistance to changes in its rotation rate. It is all the floor-foot forces act. It moves along a path during gait
the rotational analog of mass. It is particularly important and produces a characteristic pattern. GRFs, GRFVs, and
in analyzing fast-moving parts of the body during the gait CoPs that are characteristic of normal walking appear later
cycle, such as the swinging limb. in the chapter as Figures 14–10, 14–11, and 14–12.
Using the link segment model, we can determine what A moment of force, usually referred to simply as moment,
forces and moments have been acting at each joint (e.g., the is the same as torque, which was introduced in Chapter 1. It is
ankle) in order that the foot with that particular mass defined as the product of the force (usually muscle) and the
and moment of inertia moved with those particular linear perpendicular distance from its action line to the joint center.
and angular accelerations. Now we simply progress up the It can be thought of as the tendency to turn.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 530

530 SECTION 5 ■ Integrated Function

Continuing Exploration 14-1: If we refer to Figure 14–7 with reference to the foot seg-
ment, equation (1) says that that the sum of all horizontal
Understanding How the Kinetics of Gait forces must equal the product of the mass of the foot and
Are Studied its acceleration in the horizontal direction. Equation (2)
The three equations on which the solutions are based for says that the sum of all vertical forces must equal the
a two-dimensional analysis are simple: for each segment, product of the mass of the foot and its acceleration in the
the following three are applied: vertical direction. Equation (3) says the sum of the mo-
ments about any designated center (we are choosing the
∑ Fx = max (1) center of mass) must equal the product of the moment of
∑ Fy = may (2) inertia (which can be visualized as the resistance to
∑ M0 = I␣ (3) rotation) and the angular (rotatory) acceleration of the
where segment. Figure 14–7 shows that there are three things
we do not know: the horizontal force at the ankle (E),
∑ means “the sum of all of the” the vertical force at the ankle (B), and the moment around
Fx = forces in the designated x direction, in this case the ankle (Ma). Foot-to-floor forces are shown as A
horizontal, in newtons (N) and D, and are derived from a force plate. There is no
m = mass of the segment, derived from anthropometric moment about the free end of the segment.
tables, in kilograms (kg) Because we have three equations, we can solve for three
ax = acceleration of the center of mass in the x direction, unknowns but we cannot calculate the muscle moments on
derived from position and time data, in meters per opposite sides of the joint if there is co-contraction. Apply-
second squared (m/sec2) ing these three equation results in numbers for the
Fy = forces in the designated y direction, in this case horizontal force of the tibia on the ankle, the vertical
vertical, in N force of the tibia on the ankle, and the moment at the
ankle. In other words, we are finding out what forces and mo-
ay = acceleration of the center of mass derived from
ments had to have been acting at the ankle in order that the foot
position and time data, in m/sec2
with that particular mass and moment of inertia move with
M0 = moment about selected point 0, the center of mass, those particular linear and angular accelerations. In this ex-
in newton-meters (N·m), largely attributable to mus- ample30 the downward force on the ankle (B in Fig. 14–7)
cle activity, with ligaments, tendons, joint capsules, was about 500 N, the horizontal force pushing backward
and bony components involved to a lesser extent on the ankle (E) was about 100 N, and the net moment be-
I = moment of inertia, a measure of an object’s resist- ing caused by the plantarflexors tending to plantarflex the
ance to changes in its rotation rate. It is the rotational foot (Ma) was about 80 N. Now if we were continuing the
analog of mass. It is derived from anthropometric calculation we would simply progress up the body seg-
tables and expressed in kg·m2 ment by segment and solve for the lower leg, then the
α = angular acceleration of segment derived from seg- thigh, then the trunk. Larger numbers of segments and
ment position and time data, expressed in radians per three-dimensional analyses are more complex than this,
second squared (rad/sec2). A radian is about 57°. but the principles are the same.

Horizontal component of
tibia on ankle joint
Ankle joint
E
Moment (Ma) around Vertical acceleration Figure 14-7 Diagram of the foot segment in stance
ankle joint Rotatory acceleration shown with proximal joint (ankle) and all forces and
Vertical component of Horizontal acceleration moments acting on the segment. No moment is present at
tibia on ankle joint free end of the segment. Known forces and their location
A
B are the forces of the floor on the foot (A and D) and the
Center of mass of foot Vertical component of force caused by the foot’s mass (C); moments that can be
Force of mass on C force of floor on foot calculated about the center of mass are moments caused by
foot A and D, and the foot’s moment of inertia. Unknown forces
are joint reaction forces on the ankle (B and E), and the
D unknown moment is the net moment Ma at the ankle joint
Horizontal component of being caused by muscles, in this case, largely the ankle
force of floor on foot plantarflexors.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 531

CHAPTER 14 ■ Gait 531

Internal moments are moments generated by the mus- a joint is the product of the net moment and the net
cles, joint capsules, and ligaments to counteract the external angular velocity across the joint.35 If both are in the same
forces acting on the body. External forces such as the direction (flexors flexing, extensors extending, for example),
ground reaction forces produce external moments about positive work is being accomplished by energy generation.
the joints. For example, when the weight is on the heel at The most important phases of power generation and
initial contact (look at Fig. 14–11) an internal ankle dorsi- absorption during walking have been designated by joint
flexor moment produced by anterior tibial muscles will (H = hip, K = knee, A = ankle) and plane (S = sagittal, F =
oppose the external plantarflexion moment caused by the frontal, T = transverse).36
GRFV that is tending to plantarflex the foot. We will use Power absorption is accomplished when muscles per-
the internal moment convention in this chapter. For ease of form a lengthening (eccentric) contraction. They do nega-
reading, the word moment will refer to an internal moment tive work and reduce the energy of the body. If joint motion
unless otherwise indicated. and moment are in opposite directions, negative work is
Authors use different moment conventions to display being performed through energy absorption.
internal moments in figures, but they indicate the direction
of the moment by the words flexor/extensor, abductor/adductor, Electromyography
or internal/external rotator. The moment profiles presented
in this chapter display the following internal moments Electromyography (EMG) is a technique for evaluating
as positive: ankle plantarflexor, knee extensor, and hip and recording the activation signal of muscles. EMG is
extensor. performed using an instrument called an electromyograph,
Energy is the capacity to do work, and both work to produce a record called an electromyogram, in which
and energy are expressed in the same units, joules (J). the electrical activity generated by a muscle is recorded.
Work is performed by the application of force, which Electrodes may be placed on the skin surface or inserted
produces accelerations and decelerations of the body and into the muscle. There is a great deal of information about
its segments. Muscles use metabolic energy to perform EMG, the varieties of techniques that can be used, and the
mechanical work by converting metabolic energy into patterns obtained during gait.35,37–41 EMG is often used in
mechanical energy. The main objective of locomotion is conjunction with force plates, goniometry, and/or motion
to move the body through space with the least expendi- analysis systems to link the muscle activity with other events
ture of energy. during the gait cycle. The EMG record provides informa-
The overall metabolic cost incurred during locomotion tion about the time when particular muscles are acting
may be measured by assessing the body’s oxygen consump- and the relative level, or profile, of their activity. It
tion (VO2) per unit of distance traveled. If a long distance is does not tell why the muscles are acting or how much
traveled but only a small amount of oxygen is consumed, the force the muscles are generating. The reader is encour-
metabolic cost of that particular gait is low. Approximately aged to follow the developing literature on muscle
32% of maximum oxygen consumption is needed by 20- to function that is derived from elaborate mathematical sim-
30-year-olds to walk at a comfortable speed; this rises to ulations involving modeling precise muscle geometry and
48% for 75-year-olds32 and is even higher when a chronic anthropometrics.42–44 Although this work has just begun,
condition such as stroke is present.33 Metabolic equiva- it is already challenging conventional assumptions about
lents (METs) are also used to express energy cost of the the function of muscles.
activity as multiples of the resting metabolic rate. Oxygen
consumption for a person walking at 4 to 5 km/hour
averages 100 mL/kg body weight per minute, typically
2.5 to 4 METS. The greatest efficiency is attained when
the least amount of energy is necessary to travel a unit of CASE APPLICATION
distance. When asked to walk at a comfortable speed, peo- Effects on Time and
ple choose the speed at which they are most efficient, and if
Distance Gait
case 14–1
the speed of walking increases above this, the energy cost
per unit of distance walked increases.32 As the speed of walk- Variables
ing decreases below self-selected walking speed, the energy
cost increases. Both passive exchange of potential and Ms. Brown walks with a speed of 0.20 m/sec and a cadence
kinetic energy and elastic energy utilization are responsible of 25 steps per minute. It is evident even on visual inspec-
for the efficiencies,34 and these are most effective near the tion that double-support time is considerably longer than
self-selected speed. normal on both steps. Stance phase is more than 60% of the
Power generation is accomplished when muscles gait cycle on her affected side, but she spends an even
shorten (concentric contraction). They do positive work and greater proportion of time in stance on the unaffected side.
add to the total energy of the body. Power, expressed in Her left step length (unaffected side) is shorter than her
watts, is the work or energy value (in joules) divided by right (affected side), but her stride lengths are equal. Why?
the time over which it is generated. The power of muscle If you understand that a person walking in a straight line
groups performing gait is calculated through an inverse must have equal stride lengths but may have unequal step
dynamic approach. The power generated or absorbed across lengths, you understand the concepts of steps and strides.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 532

532 SECTION 5 ■ Integrated Function

CHARACTERISTICS OF NORMAL GAIT and the specific methodologies used and will not be
included here.
The approximate range of motion (ROM) needed in nor-
Time and Distance Characteristics mal gait and the time of occurrence of the maximum flexion
Means and standard deviations of walking speed, stride and extension positions for each major joint may be deter-
length, and step cadence appear in Table 14–1. The values mined by examining the joint angle profiles in Figure 14–8.
were derived from the classic work of Finley and Cody,18 The standard deviation bars (dotted lines) around the mean
who surreptitiously measured the gait of 1,100 pedestrians, profiles (solid lines) give an indication of how much person-
from Kadaba and colleagues,19 Oberg and colleagues,20 and to-person variation exists, demonstrating that 67% of sub-
colleagues of Ranchos Los Amigos National Rehabilitation jects’ values fell within the range shown. Results reported in
Center16 who obtained gait laboratory measurements. gait studies vary with age, gender, and walking speed of sub-
jects and with the method of analysis. Data presented here
were derived from three-dimensional analyses.35 For sim-
plicity, the mean value shown in the figures will be referred
Concept Cornerstone 14-1 to in the text, taken to the nearest 5°, and, to remind the
reader that these are not fixed values, the “approximately”
Normative Values for Time sign (~) will be used. In the anatomical position, the hip,
and Distance Gait Variables knee, and ankle are at approximately 0°. Flexion for the hip
and knee and dorsiflexion for the ankle are given positive
It is helpful in clinical practice to keep approximate time
values, and extension and plantarflexion are given negative
and distance measures in mind. From Table 14–1 we can
values.
see that the length of a stride is usually between a meter and
In Figure 14–8, it can be seen that the hip achieves max-
a quarter (1.25 m) and a meter and a half (1.50 m);
imum flexion (~+20°) around initial contact at 0% of the
the speed of walking is approximately this stride length
gait cycle and reaches its most extended position (~–20°) at
(one stride) per second, and step cadence is approximately
about 50% of the gait cycle, between heel-off and toe-off.
120 per minute, or two steps per stride and one stride
The knee is straight (0°) at initial contact and nearly
per second.
straight again just before heel-off at 40% of the gait cycle.
During the swing phase, the knee reaches its maximum
flexion of ~+60° at ~70% of the gait cycle. Note also that a
small knee flexion phase occurs at 10% of the gait cycle
Sagittal Plane Joint Angles and peaks at ~+15°. The ankle reaches maximum dorsiflex-
Sagittal and frontal plane kinematics and kinetics have been ion of ~+7° at approximately heel-off at about 40% of the
reasonably well described, but transverse plane data are gait cycle and reaches maximum plantarflexion (–25°) at
inconsistent and dependent on variations in joint positions toe-off (60%).

Table 14–1 Normative Values for Time and Distance Variables

CHARACTERISTIC MALE: MEAN (SD) FEMALE: MEAN (SD) SOURCE
Speed of walking (m/sec) 1.37 (0.22) 1.23 (0.22) Finley and Cody15
1.37 (0.17) 1.32 (0.16) RLA8
1.22–1.32* 1.10–1.29* Oberg et al17
1.34 (0.22) 1.27 (0.16) Kadaba et al16
Length of one stride (m) 1.48 (0.18) 1.27 (0.19) Finley and Cody15
1.48 (0.15) 1.32 (0.13) RLA8
1.23–130* 1.07–1.19* Oberg et al17
1.41 (0.14) 1.30 (0.10) Kadaba et al16
Step cadence (steps/min) 110 (10) 116 (12) Finley and Cody15
111 (7.6) 121 (8.5) RLA8
117–121* 122–130* Oberg et al17
112 (9) 115 (9) Kadaba et al16

*Range of means.
RLA, Ranchos Los Amigos.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 533

CHAPTER 14 ■ Gait 533

JOINT ANGLES
SAGITTAL PLANE

10-20˚
0˚
hyperextension
flexion
20˚ 15˚
flexion flexion
5˚ 0˚
15˚ flexion
0˚ flexion

5˚ 5˚
Plantar Dorsal 0˚
0˚ flexion flexion
Initial contact Foot flat Midstance Heel off
0% 10% 30% 40%

10-20˚ 20˚
30˚ 30˚
hyperextension flexion
flexion flexion

30˚ 60˚
flexion flexion
30˚
flexion 0˚

20˚ 10˚
Plantar Plantar 0˚ 0˚
flexion flexion
Toe off
Acceleration Midswing Deceleration
60%

HIP KNEE ANKLE

30 70
flex

10
dorsi
flex

20 60

10 50
DEGREES
DEGREES

0
DEGREES

40
0
30 -10
-10
20
plant

-20
-20
ext

ext

10
-30 0 -30
20 40 60 80 100 20 40 60 80 100 20 40 60 80 100

% Gait Cycle % Gait Cycle % Gait Cycle

Figure 14-8 Joint angles in degrees at the hip, knee, and ankle in the sagittal plane. The dotted lines in the angle diagrams
represent the standard deviation values, and the solid lines represent the mean values. (Joint angle diagrams redrawn from
Winter DA, Eng JJ, Isshac MG: A review of kinetic parameters in human walking. In Craik RL, Otis CA [eds]: Gait Analysis:
Theory and Application, pp 263-265. St. Louis, MO, Mosby-Year Book, 1994, with permission from Elsevier.)
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 534

534 SECTION 5 ■ Integrated Function

Concept Cornerstone 14-2 Ground Reaction Force and Center

of Pressure
Hip, Knee, and Ankle Range of Motion
The ground reaction forces have a typical pattern from ini-
Needed for Normal Walking tial contact to toe-off (Fig. 14–10).31 In the vertical direc-
For normal walking, we need a hip ROM from approxi- tion, the magnitudes are low at first but increase to values
mately 20° of extension to 20° of flexion, a knee range that are greater than body weight both in early stance and
from straight (0°) to 60° flexion, and an ankle range from again in late stance, with lower values at midstance. The
25° of plantarflexion to 7° dorsiflexion. If these joint first hump is related to the weight acceptance when the
ranges are not available, a gait pattern would be expected body’s downward velocity is being slowed (upward accelera-
to show considerable deviation from the norm. tion of the body’s mass). The second hump is due to push-
off and shows that the body’s center of mass is being accel-
erated upward to increase its upward velocity. In the
anteroposterior direction, the GRF is directed posteriorly
Frontal Plane Joint Angles against the foot that is making initial contact and prevents
the foot from slipping forward. It reaches a maximum mag-
During the first 20% of stance, the pelvis or the contralat- nitude of about 20% of body weight. At midstance, the force
eral side drops about 5°, which results in adduction of becomes neutral, but as the body enters the second half of
the hip (Fig. 14–9). The hip abducts smoothly to about 5° of stance phase, the vector is directed anteriorly against the
abduction, peaking about toe-off, then returns to neutral foot, enabling the person to push off. The mediolateral
at initial contact. The knee remains more or less neutral, forces are small in magnitude and are variable across indi-
except for a brief abduction peaking at about 7° in viduals. Figure 14–11 shows the GRFVs in the sagittal plane
midswing, and then returns to neutral.27 From the figure, it relative to the joint positions in early stance phase. You can
can be seen that the ankle complex everts from about 5° of see that the GRFV at initial contact passes behind the ankle
inversion to 5° of eversion in early stance and inverts about (tending to plantarflex it) but in front of the knee and hip
15° during push-off. (tending to extend them). In early stance, the GRFV passes
behind the knee (tending to flex it), then moves ahead of it
until late stance. About midstance (Fig. 14–12), the GRFV
CASE APPLICATION passes behind the hip joint and remains there until swing
Effects on Joint Angle phase begins.
Patterns
case 14–2
An examination of videotaped joint angles reveals that
The single point on the foot at which the resultant sur-
face pressure may be considered to be acting is called the
center of pressure (CoP). This point is the starting point for
Ms. Brown has no knee flexion phase in early stance, and the GRFV. The CoP of the foot on the supporting surface
she tends to fully extend her knee in midstance. She has produces a characteristic pattern (Fig. 14–13). The pattern
minimal ability to dorsiflex her ankle and has difficulty clear- for normal individuals during barefoot walking differs from
ing the floor with her affected limb as a result of poor dorsi- the patterns in which various types of footgear are used.45 In
flexion and because she does not bend her knee more than a barefoot walking, the CoP starts at the posterolateral edge
few degrees during swing phase. Instead, she tends to lift her of the heel at the beginning of the stance phase and moves
pelvis (“hike”) to clear her foot during swing on the affected in a nearly linear manner through the midfoot area, remain-
side. Her affected hip does not extend beyond neutral. How ing lateral to the midline, and then moves medially across
do Ms. Brown’s joint angle profiles vary from the normal in the ball of the foot with a large concentration along the
stance and swing phases? metatarsal break. The CoP then moves to second and first
toes during late stance.

JOINT ANGLES
FRONTAL PLANE
HIP KNEE ANKLE

12.5 5.0
15.0 Figure 14-9 Joint angles in
adduction
adduction

inversion

10.5
2.5 degrees at the hip, knee, and ankle
ANGLE (deg)
ANGLE (deg)
ANGLE (deg)

7.5
0.0 10.0 in the frontal plane. (Redrawn from
5.0
2.5 -2.5
Winter DA, Eng JJ, Isshac MG:
5.0
A review of kinetic parameters in
abduction
abduction

0.0
eversion

-5.0 human walking. In Craik RL, Otis

-2.5 0.0
-5.0
-7.5 CA [eds]: Gait Analysis: Theory and
-7.5 -10.0 -5.0 Application, pp 263-265. St. Louis,
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100
% Gait Cycle
MO, Mosby-Year Book, 1994, with
% Gait Cycle % Gait Cycle permission from Elsevier.)
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 535

CHAPTER 14 ■ Gait 535

1.4 Continuing Exploration 14-2:

1.2
Why Don’t Hip, Knee, and Ankle Moment
Vertical force

1
0.8 Figures Look Familiar to Me?
0.6 Sometimes when we are reading journal articles that in-
0.4 clude figures and explanations of moments at the hip,
knee, or ankle during walking, the shapes seem totally
0.2
unfamiliar and the explanations do not make sense. In fact
0
they often feel “reversed,” which, in fact, is exactly what
0 20 40 60 80 100 has happened. There are two ways of reporting moments.
% stance Here we have chosen to use “internal” moments, which
refer to the moments produced by the muscles and other struc-
tures around the joint that had to have been acting to move
0.08 body parts (with their known characteristics) in the direc-
Mediolateral force

tions and with the accelerations that we observe with

0.04
motions capture systems. However, some authors use
0 “external” moments, which refer to the moments pro-
20 40 60 80 100 duced by the external forces acting around the joint that
–0.04 must be acting in order that the body parts (with their
–0.08
known characteristics) were moving in the way they were.
% stance
For example, if the internal moment of the knee in early
stance was found to be an extensor moment, being caused
by the quadriceps muscle, the external moment would be
0.4 a flexor moment, caused largely by the GRFV passing
Anteroposterior force

behind the knee “trying to” flex it. Similarly, an internal

0.2 abductor moment is an external adductor moment—but
the magnitude is exactly the same. Therefore, internal
0
moments must match the magnitude of the external
20 40 60 80 100 moments, but in the opposite direction, to result in the
–0.2
% stance positions and motions that we observe.
–0.4

Figure 14-10 Ground reaction forces in newtons as a propor-

tion of body weight over stance phase. Positive values represent Sagittal Plane Moments
upward, lateral, and posterior forces acting on the foot. (Adapted
and redrawn from Hunt AE, Smith RM, Torode M, et al: Inter- Before concerning ourselves with the individual sagittal
segment foot motion and ground reaction forces over the stance phase plane moments about the hip, knee, and ankle, let us con-
of walking. Clin Biomech [Bristol, Avon] 16:592, 2001.) sider the concept of the support moment, which will help

GRFV
GRFV GRFV Flexor
moment
20˚ 15˚ 0˚
Extensor Extensor
flexion flexion flexion
moment moment

Extensor
moment 5˚
Flexor
0˚ 15˚ moment flexion
flexion
Flexor
moment
Plantar
0˚ Dorsiflexor 5˚ Dorsiflexor 5˚
flexor
moment Plantar moment Dorsal
moment
flexion flexion
CoP CoP CoP
Initial contact Foot flat Midstance
Figure 14-11 Diagram of joint positions with center of pressure (CoP), ground reaction force vector
(GRFV), and internal net moments of force for gait cycle events of initial contact, foot flat, and midstance.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 536

536 SECTION 5 ■ Integrated Function

GRFV GRFV GRFV

Flexor Flexor
moment moment Flexor
moment
0˚ 20˚
flexion extension
30˚
flexion
Flexor 5˚ Flexor Extensor
moment flexion moment 0˚ moment
Plantar
flexor 25˚
5˚ moment Plantar
Plantar Dorsal Plantar flexion
flexor flexion flexor 0˚ (Toes) 60˚
Flexor hyperextension
moment moment Flexor
moment moment
CoP CoP (toes) CoP (toes)
Midstance Heel off Toe off

Figure 14-12 Diagram of joint positions with center of pressure (CoP), ground reaction force vector
(GRFV), and internal net moments of force for gait cycle events of midstance, heel-off, and toe-off.

A B C D
Figure 14-13 A center of pressure (CoP) pathway is shown by the position of the black dot at initial
contact (A), at foot flat (B), just before heel-off (C), and just before toe-off (D).

us to understand the moment patterns through the gait support provided by the ankle plantarflexors until they
cycle. The sum of the hip, knee, and ankle moments keeps become the only support in most of late stance. The support
the leg from collapsing during the stance phase. That is, moment changes from a net extensor to a net flexor mo-
for most of stance phase, the sum of all moments acting at ment in late stance (55% to 60% of the gait cycle), which
the hip, knee, and ankle is a positive or extensor moment initiates swing. In late swing, a small net extensor moment
(Fig. 14–14). Winter5 called this total limb synergy a sup- appears again, to assist in the final positioning of the limb
port moment. He found the extensor support moment to for initial contact.5,45
be consistent for all walking speeds for both normal indi- A clinically important feature of the support moment is
viduals and persons with disabilities. In Figure 14–14, no- that as long as the moments add up to be extensors, the body
tice that hip extension provides all of the positive moment can vary how it accomplishes its support. For example, if the
in early stance, but it is soon joined by a knee extensor mo- ankle plantarflexors are weak, the hip extensors and/or the
ment. (Remember that the moment identifies only the knee extensors can compensate. If contraction of the knee
“supportingness” of the muscle group; it does not mean extensors causes pain, the hip extensor and/or ankle plan-
that muscle shortening or lengthening is occurring. To tarflexors can contract harder.
find out if the muscle is shortening or lengthening, one The support moment helps us to understand the joint
needs to consult the joint angle profiles: Even if there is no moment profiles for each joint. The sagittal plane moment
change in joint angle a moment may be present. In normal profiles for the hip, knee, and ankle are shown with GRFVs
gait, the knee extensors are first lengthening and then in Figure 14–15, and with joint angle profiles and muscle
shortening.) As stance phase proceeds, there is increasing work in Figure14–16.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 537

CHAPTER 14 ■ Gait 537

Moment of Force-Natural Cadence (N = 19)

CASE APPLICATION

2 Sum of hip, knee, and ankle Difficulty in Developing an

Adequate Support
case 14–3
EXT

1 SUPPORT Moment
MOMENT
Ms. Brown has low levels of activation of her ankle plan-
0 tarflexors, knee extensors, and, to a lesser extent, hip
extensors. This will make it difficult for her to develop an
Moment / Bodymass (N.m / KG )

1 Hip extensors HIP adequate support moment. To gain knee stability she
tends to thrust her knee backward. We know that she has
EXT

0 smaller deficits at the hip than at the ankle, and so we

would encourage hip extension in early stance to assist
-1 with the support moment, as well as trying to stimulate
the knee extensors. During this early poststroke period,
1 Knee extensors we would expect some natural increase in force-generating
KNEE
capability of the muscles (strength) and we would be pre-
EXT

0
pared to take advantage of it by strengthening the muscles
during functional movements. If over time she is unable to
provide enough support, an ankle orthosis would provide
-2 a passive extensor moment, but there may be energy costs
to doing this.
2 Ankle plantar flexors
PLANTAR

1 ANKLE

Continuing Exploration 14-3:

0
TOE OFF = 60% Kinematics and Kinetics of the Foot
and Ankle
100
20

40

60

80
0

Specific descriptions of the biomechanics of the foot

% of stride
during gait have been hindered by the complexity of the
Figure 14-14 Typical pattern of sagittal plane moments at the composite movements created by the ankle, subtalar, and
hip, knee, and ankle, shown with their algebraic sum, the support transverse tarsal joints (and their noncardinal axes). These
moment. (Redrawn from Winter DA: The Biomechanics and Motor kinematics are referred to as pronation (a composite
Control of Human Gait: Normal, Elderly and Pathological [ed. 2]. of dorsiflexion, eversion, and abduction) and supination (a
Waterloo, Ontario, Waterloo Biomechanics, 1991, with permission from composite of plantarflexion, inversion, and adduction).
David A. Winter.)
Usually, the foot movement is described as consisting of
pronation early in stance, followed by progressive supina-
tion. However, three-dimensional analyses of the lower
Frontal Plane Moments leg and foot, modeling the foot as a rearfoot segment
and a forefoot segment, have provided more insight into
Frontal plane moments are shown in Figure 14–17 and with its behavior during stance.46,47 Although kinematics and
joint angle profiles and muscle work in Figure 14–18. Large kinetics are highly variable and depend on subject, foot
abduction moments of similar shapes occur at the hip and position, terrain, and method of analysis, some general
knee, and a smaller one occurs at the ankle. These are pro- features are thought to occur. The rearfoot segment
vided largely by ligament forces across the knee and ankle everts with respect to the lower leg early in stance and in-
joints, and are necessary as the center of mass of the body is verts during push-off. In the transverse plane, the rearfoot
considerably medial to the point of support on the foot. segment externally rotates, (abducts) in early stance, fol-
There appears to be some active component at the knee, lowed by internal rotation (adduction) through push-off.
however (either muscular or passive spring related), because The forefoot segment adds considerably to dorsiflexion
some power generation will be seen in the section on frontal during early stance.
plane powers.
Text continued on page 540
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 538

538 SECTION 5 ■ Integrated Function

Internal Moments
SAGITTAL PLANE
GRFV
GRFV GRFV
Flexor
Extensor Extensor moment
moment moment

Extensor Flexor
moment moment
Flexor
moment Dorsiflexor
Dorsiflexor
moment Plantar
moment flexor
moment
CoP CoP CoP
A Initial contact Foot flat Midstance

GRFV GRFV GRFV

Flexor Flexor
moment moment Flexor
moment

Flexor Flexor Extensor

moment moment moment

Plantar
Plantar Plantar
flexor
flexor flexor Flexor Flexor
moment
moment moment moment moment
CoP CoP (toes) CoP (toes)
B Midstance Heel off Toe off

HIP KNEE ANKLE

Sagittal
Joint Moment
0.75
1.25
extensor
extensor

1.75
plant

0.50
0.75
1.25
0.25 0.25
N.m/kg

N.m/kg
N.m/kg

0.75
-0.25 0.00

-0.25 0.50
-0.75
flexor

dorsi
flexor

-1.25 -0.50 0.25

0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

C % Gait Cycle % Gait Cycle % Gait Cycle

Figure 14-15 Patterns of internal moments in the sagittal plane at the hip, knee, and ankle with center of pressure (CoP)
and ground reaction force vectors (GRFVs). The dotted lines represent the standard deviations, and the solid lines represent
the mean values. (Diagrams of internal moments redrawn from Winter DA, Eng JJ, Isshac MG: A review of kinetic parameters in
human walking. In Craik RL, Otis CA [eds]: Gait Analysis: Theory and Application, pp 263-265. St. Louis, MO, Mosby-Year Book,
1994, with permission from Elsevier.)
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 539

CHAPTER 14 ■ Gait 539

Muscle Activity

GLUTEUS MAXIMUS

20 40 60 80 100

Sagittal Sagittal
Joint Angle Joint Moment
MEDIAL HAMSTRING
flexion

30
1.25

extensor
20
0.75
20 40 60 80 100
10
DEGREES

0.25 ILIOPSOAS
HIP

Nm/kg
0
-0.25
-10
extension

20 40 60 80 100
flexor
-20 -0.75

-30 -1.25 VASTUS LATERALIS

20 40 60 80 100 0 20 40 60 80 100

70 0.75 20 40 60 80 100
flexion

extensor

60
0.50

Normalized Electromyography (%)

50 MEDIAL
GASTROCNEMIU
DEGREES

40 0.25
Nm/kg

KNEE
30 0.00
20 40 60 80 100
extension

20
-0.25
flexor

10

0 -0.50
20 40 60 80 100 0 20 40 60 80 100 LATERAL HAMSTRING

Sagittal
flexion
dorsi

20 40 60 80 100
10 1.75
plantar

SOLEUS
DEGREES

0 1.25
ANKLE
Nm/kg

-10 0.75
extension
plantar

0.50 20 40 60 80 100
-20
dorsi

-30 0.25
20 40 60 80 100 0 20 40 60 80 100 TIBIALIS ANTERIOR

% Gait Cycle % Gait Cycle

20 40 60 80 100

% Gait Cycle

Figure 14-16 Joint angles and net joint moments in the sagittal plane, and EMG profiles of representatives of major contributors to
joint moments of hip, knee, and ankle during adult gait. (Angle and moment profiles redrawn from Winter DA, Eng JJ, Isshac MG: A review
of kinetic parameters in human walking. In Craik RL, Otis CA [eds]: Gait Analysis: Theory and Application, pp 263-265. St. Louis, MO, Mosby-
Year Book, 1994, with permission from Elsevier. Muscle activity redrawn from Winter DA: The Biomechanics and Motor Control of Human Gait:
Normal, Elderly and Pathological [ed. 2]. Waterloo, Ontario, Waterloo Biomechanics, 1991, with permission from David A. Winter. Iliopsoas muscle
activity redrawn from Bechtol CO: Normal Human Gait. In American Academy of Orthopaedic Surgeons: Atlas of Orthotics, p 141. St Louis, MO,
CV Mosby, 1974, with permission from Elsevier.)
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 540

540 SECTION 5 ■ Integrated Function

INTERNAL MOMENTS
FRONTAL PLANE
HIP KNEE ANKLE
Figure 14-17 Patterns of inter-
0.25
nal moments in the frontal plane

abductor
1.50
abductor

everrtor
0.95 at the hip, knee, and ankle. The
1.25
0.75 dotted lines represent the standard
1.00 deviation, and the solid lines indi-

(N.m/kg)
0.15

(N.m/kg)
0.55
(N.m/kg)

0.75 cate the mean values. (Redrawn

0.35 from Winter DA, Eng JJ, Isshac
0.50
adductor 0.05 MG: A review of kinetic parameters
adductor

0.25 0.15

invertor
in human walking. In Craik RL,
0.00 -0.05
Otis CA [eds]: Gait Analysis: Theory
-0.25
0 20 40 60 80 100
-0.25
0 20 40 60 80 100
-0.05
0 20 40 60 80 100
and Application, pp 263-265. St.
Louis, MO, Mosby-Year Book, 1994,
% Gait Cycle % Gait Cycle % Gait Cycle with permission from Elsevier.)

Sagittal Plane Powers contact, negative work is performed by the plantarflexors

Sagittal plane power profiles across the hip, knee, and ankle (A1-S) as the leg rotates over the foot during the period of
are compared with joint angle and major muscle activity pro- stance from foot flat to about 40% of the gait cycle. How-
files in Figure 14–19. A summary of the phases of power gen- ever, a small amount of positive work is done by the knee
eration and absorption during the full gait cycle appear in extensors at the beginning of this period (K2-S), extend-
Table 14–2. Table 14–3 on page 549 shows powers with joint ing the knee after foot flat.48 Positive work of the plan-
motion, GRFs, moments, and muscle activity from initial tarflexors at push-off (A2-S: late stance, ~40% to 60% of
contact to midstance, and Table 14–4 on page 550 for mid- the gait cycle) and hip flexors at pull-off (H3-S: late stance
stance to toe-off. Despite the complexity of the appearance and in early swing, ~50% to 75% of the gait cycle) in-
of this information, the important facts are straightforward. creases the energy level of the body. During this 40% to
Of greatest importance is the fact that most of the positive 60% of the gait cycle when energy is being generated
work of walking is provided by the ankle plantarflexors at from A2-S and H3-S, simultaneous absorption is occur-
push-off, the hip extensors in early stance, and the hip flex- ring by knee extensors (K3-S). In late swing, negative
ors in late stance and early swing (“pull-off”). Second, the work is performed by the knee flexors (K4-S) as they work
knee extensors normally provide very little power, and act eccentrically to decelerate the leg in preparation for ini-
largely as absorbers. To make it possible to compare subjects, tial contact.
power values have been normalized by dividing the power in At slow and normal speeds of walking in healthy subjects,
watts by the subjects’ weight in kilograms. the hip flexors and extensors contribute about 25% of the
In the sagittal plane (see Fig. 14–19), a burst of positive total concentric work.49 The ankle plantarflexors contribute
work (energy generation) occurs as the hip extensors about 66%, and the knee extensors about 8%.49 Clearly, the
contract concentrically during early stance (H1-S), ankle plantarflexors are of primary importance in walking
while the knee extensors perform negative work (energy (see “Continuing Exploration: On Rockers and Push-Off”).
absorption) by acting eccentrically (K1-S) to control knee We can think of the ankle and hip muscles as primarily con-
flexion during the same period.48 Following a tiny period cerned with the function of generation of energy in walking
of unnamed negative work by the dorsiflexors at initial and the knee muscles as absorbers of energy.4,17

Table 14–2 Summary of Major Phases of Power Generation and Absorption During
Full Gait Cycle
GENERATION (CONCENTRIC CONTRACTION) ABSORPTION (ECCENTRIC CONTRACTION)
Name of Major Muscle Occurrence in Name of Major Muscle Occurrence in
Power Burst or Group Gait Cycle Power Burst or Group Gait Cycle
H1-S Hamstrings, Early stance H2-S Psoas, rectus Midstance
gluteals
K2-S (small) Quadriceps Early stance K1-S Hamstrings Very early stance
H3-S Psoas, rectus Late stance, K3-S Rectus femoris Late stance
“pull-off”
A2-S Plantarflexors Late stance, K4-S Hamstrings Late swing
“push-off”
A1-S Plantarflexors Early and midstance
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 541

CHAPTER 14 ■ Gait 541

Muscle Activity

GLUTEUS MEDIUS

Frontal Frontal
Joint Angle Joint Moment
12.5 1.50
adduction

abductor
10.5 1.25

7.5 1.00
20 40 60 80 100

Normalized EMG (%)

ANGLE (deg)

5.0

(N.m/kg)
0.75
HIP

GLUTEUS MEDIUS
2.5
0.50
0.0
abduction

0.25

adductor
-2.5
0.00
-5.0
-7.5 -0.25
0 20 40 60 80 100 0 20 40 60 80 100
20 40 60 80 100

5.0 0.95 ( largely ligamentous)

abductor
adduction

2.5 TENSOR FASCIA LATAE

0.75

Normalized EMG (%)

0.0 0.55
ANGLE (deg)

(N.m/kg)
KNEE

-2.5 0.35

-5.0 0.15
abduction

adductor

-7.5 -0.05 20 40 60 80 100

-10.0 -0.25
0 20 40 60 80 100 0 20 40 60 80 100
PERONEUS LONGUS

Electromyography (%)

0.25
evertor
inversion

15.0

10.0 0.15
ANGLE (deg)
ANKLE

Normalized

20 40 60 80 100
(N.m/kg)

5.0
0.05
eversion

invertor

0.0

-5.0 -0.05
0 20 40 60 80 100 0 20 40 60 80 100
20 40 60 80 100
% Gait Cycle % Gait Cycle % Gait Cycle

Figure 14-18 Joint angles and net joint moments in the frontal plane, and EMG profiles of representatives of major contributors to
joint moments of hip, knee, and ankle during adult gait. (Angle and moment profiles redrawn from Winter DA, Eng JJ, Isshac MG: A review
of kinetic parameters in human walking. In Craik RL, Otis CA [eds]: Gait Analysis: Theory and Application, pp 263–265. St. Louis, MO, Mosby-
Year Book, 1994, with permission from Elsevier. Muscle activity redrawn from from Winter DA: The Biomechanics and Motor Control of Human
Gait: Normal, Elderly and Pathological [ed. 2]. Waterloo, Ontario, Waterloo Biomechanics, 1991, with permission from David A. Winter.)
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 542

542 SECTION 5 ■ Integrated Function

CASE APPLICATION Unless gait velocity is increasing, energy has to be sys-

tematically removed. The knee accomplishes this at K1-S
Effects of an Ankle-Foot (knee flexion with a knee extensor moment) and K3-S
Orthosis
case 14–4
Recall that Ms. Brown hiked up her affected side (lifted hip
(small knee extensor moment in late stance occurring while
the knee is flexing quite quickly). The latter may represent
inefficiency in gait.52 K4-S (knee flexor absorption with
and pelvis) in order to clear her foot during swing phase. This knee extension) occurs before initial contact. Note that, as
has been shown to have a serious energy cost51 as the full with all absorption phases, that is, when eccentric muscle activ-
weight of the upper body is raised and lowered, and there are ity is dominating, the dominating moment is opposite to the
no opportunities for savings by kinetic-potential exchange. movement that is occurring. In this case, a flexor moment is
Because the inability to adequately dorsiflex her foot ap- dominating (it is the net moment) while knee extension is
peared to be part of the problem, a light ankle-foot orthosis occurring. The knee flexors are acting eccentrically, and
was prescribed, which held her ankle in a neutral position the knee is being pulled into extension. A simpler example
during swing phase. The orthosis also assisted with her in- occurs when you squat. In this case, a knee extensor mo-
ability to provide sufficient support during stance. The some- ment is dominating while knee flexion is occurring. Energy
what stiff orthosis material accomplished this by resisting is being removed from the body, which is very evident as
dorsiflexion of her ankle during stance, as her body pro- you end up in a lower position than you were; your poten-
gressed forward over her foot. She gained the ability to bend tial energy is therefore lower.
her knee in later stance, thus avoiding excessive hip hiking. Let us now examine the periods during which simulta-
Watch for possible drawbacks while you read the section on neous positive and negative work normally occur, which, if
power and work. If the orthosis does not permit any ankle excessive, represents inefficiency. The most important one
plantarflexion during push-off, she will lose the energy that occurs in late stance: Negative work done by the flexing
would otherwise be provided by plantarflexor generation. knee (K3-S) occurs concurrently with both positive work
by the hip flexors (H3-S—pull-off) and ankle plantarflexors
(A2-S—push-off). Values of K3-S that are above normal
represent excessive inefficiency and usually result in slower
walking speeds.
Continuing Exploration 14-4:
Understanding Power and Work in Gait
Examination of power plots helps us identify and under-
Frontal Plane Powers
stand the contribution of lower extremity muscles to gait, and In the frontal plane (Fig. 14–20), an initial period of
its phases. When slow or inefficient gait is a problem, absorption by the hip abductors (H1-F) is followed by two
knowledge of the sources of power enables a health practi- small bursts of positive work in the remainder of stance
tioner to assist the client to compensate for deficiencies. (H2-F, H3-F). These bursts provide fine control of the
Note that the scales used for power in the figures vary, and mediolateral position of the center of mass of the body. At
this must be taken into account when assessing the work the knee, there is a very small generation pattern during
(the area under the power curve). In Figure 14–19, identify the first half of stance (K1-F), followed by a small absorp-
the joint around which the largest amount of positive work tion (K2-F) both caused by abductor structures. Ankle
(above zero line) is performed in the sagittal plane. It is ap- power patterns are not shown as they are small and some-
parent that the burst denoted A2-S (S denoting burst in the what inconsistent.
sagittal plane) is the largest, and this occurs at the ankle. We
know that this positive work may be caused by either ankle Mechanical Energy of Walking
plantarflexor activity with plantarflexion, or else active
Walking at a constant speed is accomplished by the bursts of
ankle dorsiflexors with dorsiflexion. We know this because
positive (concentric) and negative (eccentric) muscle
positive work is occurring. A quick examination of the joint
work described above that result in small stride-to-stride
angle profile shows us that ankle plantarflexion is occurring
fluctuations in body height and variations in the velocities of
at this time so we know the burst of positive work is caused
the body parts. One way the body saves energy is by making
by the plantarflexors. The second largest burst of positive
use of passive exchange of potential and kinetic energy
power, denoted H1-S, is a hip extensor moment with hip
through the stride, a process that is much like a ball rolling
extension occurring in early stance. The third burst, H3-S,
down one hill and up another.
occurs near A2-S. It is caused by a hip flexor moment and
Figure 14–21 is a diagram of kinetic energy of the head,
hip flexion that occur at the end of stance and the begin-
arms, and trunk (bottom line), their potential energy
ning of swing. It is frequently called pull-off.
(middle), and the two added together (top).50 At initial
The knee is not of major importance in energy gener-
contact of each foot, that is at 0% and 50% of the gait
ation. However, there is a small phase, denoted K2-S, in
cycle, the body has the lowest potential energy but is
which the knee extensors extend the knee after the knee
moving the fastest, and as it moves into midstance, the
flexion phase of early stance. K2-S can be important in
potential energy rises and is exchanged for kinetic
pathologies such as cerebral palsy when other sources of
energy. Evidence of the energy exchange can be seen as
energy are not available.
the potential energy curve slopes downward following
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 543

CHAPTER 14 ■ Gait 543

Muscle Activity
GLUTEUS MAXIMUS

Sagittal Sagittal 0 20 40 60 80 100

Joint Angle Joint Power
30
MEDIAL HAMSTRING
flexion

GEN
20 2.0
HIP H1-S H3-S
DEGREES

WATTS / Kg
10
0 20 40 60 80 100
0 0 ILIOPSOAS

N O R M A L I Z E D E L E C T R O M YO G R A P H Y ( % )
extension

-10

ABS
-20 H2-S
-2.0
0 20 40 60 80 100
-30 VASTUS LATERALIS
20 40 60 80 100
70
60 60 80 100
flexion

0 20 40
KNEE
GEN

1.2 KO-S
DEGREES

50 K2-S RECTUS FEMORIS

WATTS / Kg

40 0
30
extension

20 K1-S
K4-S
ABS

10 -2.5 K3-S 0 20 40 60 80 100

0 0 20 40 60 80 100
20 40 60 80 100 LATERAL
Sagittal HAMSTRING

10 0 20 40 60 80 100
flexion

GEN

ANKLE 2.0 A2-S SOLEUS

DEGREES

0
WATTS / Kg

-10
extension

0 0 20 40 60 80 100
-20
ABS

-1.5 A1-S
-30 0 20 40 60 80 100 TIBIALIS ANTERIOR
20 40 60 80 100
% Gait Cycle

0 20 40 60 80 100
% Gait Cycle

Figure 14-19 Joint angles and joint powers in the sagittal plane, and EMG profiles of representatives of major contributors to joint
powers of hip, knee, and ankle during adult gait. (Angle profiles redrawn from Winter DA, Eng JJ, Isshac MG: A review of kinetic parameters
in human walking. In Craik RL, Otis CA [eds]: Gait Analysis: Theory and Application, pp 263-265. St. Louis, MO, Mosby-Year Book, 1994,
with permission from Elsevier. Power profiles redrawn from from Eng JJ, Winter DA: Kinetic analysis of the lower limbs during walking: What
information can be gained from a three dimensional model? J Biomech 28:753, 1995, with permission from Elsevier. Muscle activity redrawn
from Winter DA: The Biomechanics and Motor Control of Human Gait: Normal, Elderly and Pathological [ed. 2]. Waterloo, Ontario, Waterloo
Biomechanics, 1991, with permission from David A. Winter. Iliopsoas muscle activity redrawn from Bechtol CO: Normal Human Gait. In
American Academy of Orthopaedic Surgeons: Atlas of Orthotics, p 141. St Louis, MO, CV Mosby, 1974, with permission from Elsevier.)
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 544

544 SECTION 5 ■ Integrated Function

Muscle Activity

GLUTEUS MEDIUS

NORMALIZED ELECTROMYOGRAPHY (%)

Frontal Frontal
Joint Angle Joint Power
12.5
adduction

10.5

7.5
H3-F 0 20 40 60 80 100
0.75

GEN
ANGLE (deg)

5.0 H2-F
ADDUCTOR MAGNUS
HIP

2.5

WATTS / Kg
0
0.0
abduction

-2.5

-5.0

-7.5 ABS -1.50 H1-F

0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

5.0
adduction

2.5 (Ligamentous)

0.0 0.35
GEN
ANGLE (deg)

K1-F TENSOR FASCIA LATAE

KNEE

-2.5
WATTS / Kg

-5.0 0
abduction

-7.5
K2-F
ABS

-0.30
-10.0
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

Figure 14-20 Joint angles and joint powers of hip and knee in the frontal plane, and EMG profiles of representatives of major con-
tributors to joint powers of hip and knee during adult gait. (Angle profiles redrawn from Winter DA, Eng JJ, Isshac MG: A review of kinetic
parameters in human walking. In Craik RL, Otis CA [eds]: Gait Analysis: Theory and Application, pp 263-265. St. Louis, MO, Mosby-Year Book,
1994, with permission from Elsevier. Power profiles redrawn from Eng JJ, Winter DA: Kinetic analysis of the lower limbs during walking: What
information can be gained from a three dimensional model? J Biomech 28:753, 1995, with permission. Muscle activity redrawn from Winter DA:
The Biomechanics and Motor Control of Human Gait: Normal, Elderly and Pathological [ed. 2]. Waterloo, Ontario, Waterloo Biomechanics, 1991,
with permission from David A. Winter.)

CASE APPLICATION Ms. Brown was encouraged to exploit H1-S, the “push
Continuing Gait from behind” in early stance. Stronger activity of A2-S,
H1-S, and H3-S on the unaffected side would be encour-
Problems
case 14–5
It was noted that Ms. Brown tended to fully extend her
aged, especially later in stages of rehabilitation, to provide
interlimb compensation for the reduced activity on the
affected side. Gait speed would increase to the degree that
knee in midstance and then flexed her knee only a few de- these efforts are effective.
grees during swing phase. A power analysis of Ms. Brown’s
affected limb, if available, would have shown a severely re-
duced A2-S and H3-S and virtual absence of an H1-S en-
ergy burst. The first two would be apparent to the therapist Continuing Exploration 14-5:
as no firm push-off28 and no rapid hip flexion. This meant On Rockers and Push-Off
that Ms. Brown would be unable to push off strongly
(A2-S) or pull off strongly (H3-S), because both actions re- There has been some controversy concerning the terms
quire knee flexion. Every attempt was made during gait rockers and push-off. Orthopedic literature frequently
training to gain knee control in midstance, not permitting refers to three rockers as characterizing normal kinemat-
it to fully extend. Strong push-off and strong “pull-off” ics of the foot and ankle during stance phase. The first
then could be encouraged. Because the hip extensors rocker, or initial rocker, occurs from initial contact until
on the affected side were among the least affected muscles, foot flat and has a fulcrum about the heel (heel pivot). The
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 545

CHAPTER 14 ■ Gait 545

Initial contact
second rocker, or midstance rocker, is described as occur- 440 right foot
ring between foot flat and heel-off and has a fulcrum
about the ankle joint. The third rocker, or terminal
430
rocker, occurs between heel-off and toe-off (push-off), Total
with the leg rotating about the forefoot. This terminology energy
410 Midstance of HAT
is frequently used to denote abnormalities occurring in R L
particular phases, as, for example, “. . . walking with an an-
kle-foot orthosis impaired [the] third rocker”53 and does 400

not imply that there is no push-off. Although the termi-

nology is not incorrect, it is preferable that standard 380

Energy, joules
terminology be used. Potential
energy
Before kinetic link segment analysis was common, 360
Total energy of HAT
there were objections to the term push-off. It was thought Potential energy
that the second peak in the vertical ground reaction force 50 Kinetic (translational)
energy
was passive and resulted from changes in body alignment,
rather than an increasing force resulting from plan-

Initial contact, left

40 Kinetic
tarflexor contraction. However, the work of Winter54 and energy

Toe off, right

Toe off, left
others42,43,52 shows that push-off not only exists but is 30
of HAT
normally responsible for a major portion of the work of
walking. It is clear that a person with no active ankle plan- 20
tarflexion (hence no power generation at the ankle) can 0 200 400 600 800 1000
walk, however. Some compensation can be provided by Time, ms
the hip extensors in early stance (H1-S) and the hip Figure 14-21 Potential and kinetic energy levels of the head,
flexors in late stance (H3-S) from both limbs. Zajac and arms, and trunk (HAT) segment, and their sum for one stride
colleagues42,44 and other researchers have shown that in gait. (Redrawn from Winter DA, Quanbury AO, Reimer GD:
the energy produced by the soleus muscle in late stance is Analysis of instaneous energy of normal gait. J Biomech 9:253, 1976,
delivered to the trunk to accelerate it forward, but the with permission from Elsevier.)
increase in trunk energy is more than that produced by
the soleus muscle.43,44 It appears that the soleus, rectus
femoris, and gastrocnemius work synergistically to ensure
forward acceleration of the trunk. These studies show that The muscle work of gait can be simplified by anticipat-
muscles not only generate and absorb energy but, in many ing the muscle groups that must be functioning for
cases, serve to redistribute energy. specific purposes. Muscle work is logical, and the reader
already knows that two major features are going to deter-
mine its patterns: the need to provide a support mo-
ment through stance and the need to generate energy
midstance at the same times as the slope of kinetic energy to move.
rises. In this way there are great energy savings, as can Let us follow the logic of what muscle work around the
be seen in the top curve, which is the sum of potential hip, knee and ankle is expected if we consider what we know
and kinetic components. It is also apparent that if kinetic about the support moment (see Figs. 14–22 and 14–23).
energy does not match potential energy in magnitude—that The support moment is made up of some combination of
is, if the person walks much more slowly or much more hip extensors, knee extensors, and ankle plantarflexors.
quickly—conservation is reduced. The lower limbs are We already have discussed the fact that the hip extensors
larger contributors to the total energy costs than are the produce an extensor moment early in stance. Next, knee
head, arms, and trunk, but the opportunities for modifica- extensors produce a knee extensor moment, and then
tions by training are limited. flexor moments are produced at both the hip and knee be-
fore the hip and knee bend in late stance in preparation for
swing phase. While these two muscle groups are flexing the
Muscle Activity segmented system, the ankle plantarflexors take the lead
EMG studies of gait are used to augment the understand- and provide support. We should not be surprised, then, to
ing provided by a link segment analysis, to validate find activity in the hip extensors (hamstrings) early in
theoretical models that attempt to explain why muscles stance and knee extensors (quadriceps) almost immediately
are needed for certain functions, and in theoretical mod- after, followed by a smoothly increasing contraction of the
els developed to explain the muscle activity found by ankle plantarflexors (soleus, medial and lateral gastroc-
EMG.41 The EMG reported in the section that follows nemius muscles and other minor contributors) that con-
is derived from the work of Winter17 and is shown in tinues until mid-push-off and then declines and ceases at
Figures 14–22, 14–23, and 14–24. EMG with respective about 60% of the cycle.
joint angles and moments appears in Figures 14–16 and Now let us look for the muscle work that is responsible
14–18, and with repective joint angles and powers in for the main bursts of positive work, relating the muscle
Figures 14–19 and 14–20. work shown in Figures 14–22 and 14–23 to power profiles
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 546

546 SECTION 5 ■ Integrated Function

400 GLUTEUS MAXIMUS N=16 250 ADDUCTOR MAGNUS N=11

300 MEAN (ᒒV)=16.4 200 MEAN (ᒒV)=42.7

150
200
100
100
50
0 0
0

100
20

40

60

80

100
20

40

60

80
500 MEDIAL HAMSTRING N=23 250 ADDUCTOR LONGUS N=16
MEAN (ᒒV)=68.5
400 200 MEAN (ᒒV)=33.9

300 150
200 100
100 50
0 0
0

0
100
20

40

60

80

100
20

40

60

80
LATERAL HAMSTRING N=27 GLUTEUS MEDIUS N=26
400 400
MEAN (ᒒV)=54.5
300 300
MEAN (ᒒV)=29.7
200 200

100 100
Figure 14-22 Electromyo-
0 0
graphic activity profiles of major
muscle groups of the hip during
100
20

40

60

80

20

40

60

80
0

0
100

one stride of gait with probable

roles. Signals were derived from
bipolar surface electrodes as data
300 SARTORIUS N=15 300 TENSOR FASCIA LATAE N=8 normalized to means for each sub-
ject. Note absence of the profile
250 MEAN (ᒒV)=25.5 250
for the hip flexor, the iliopsoas
200 200 muscle, which is inaccessible to
MEAN (ᒒV)=34.5
150 150 surface electrodes. (Redrawn from
100 100 Winter DA: The Biomechanics
and Motor Control of Human Gait:
50 50
Normal, Elderly and Pathological
0 0 [ed. 2]. Waterloo, Ontario, Waterloo
Biomechanics, 1991, with permission
100
100

20

40

60

80
20

40

60

80

0
0

from David A. Winter.)

shown in Figures 14–19 and 14–20. In the sagittal plane, we foot. At about 40% of the gait cycle, they produce a burst
note that the hip extensors (gluteus maximus, medial of concentric activity (A2-S) ending at toe-off. A similar
hamstring, and lateral hamstring muscles) are active in sequence of first eccentric and then concentric activity
early stance (H1-S); indeed, they were serving the function occurs in the hip flexors: iliopsoas and rectus femoris
of providing support during that period. The small energy- muscle. First, they lengthen, producing an energy-absorbing
generating K2-S that peaks in early stance is reflected in contraction as the hip extends (H2-S); then the muscle force
activation of the quadriceps (vastus medialis, vastus lat- overcomes the opposing ground reaction force and begins
eralis, and rectus femoris muscles). Recall that the largest to act concentrically, causing an energy-generating “pull-off”
contribution to the work of walking comes from the ankle phase (H3-S). The iliopsoas muscle is the major hip flexor,
plantarflexors (largely soleus, medial, and lateral gas- but it is inaccessible to surface electrodes. However,
trocnemius muscles). These muscles work eccentrically the rectus femoris muscle, being the only one of the
(lengthening) from early stance until about 40% of the gait quadriceps muscle that crosses the hip, also shows hip flexor
cycle, when they are controlling the forward movement of function, and its activity in late stance correlates with that of
the tibia over the talus as the upper body passes over the iliopsoas in late stance. We can see rectus femoris activity
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 547

CHAPTER 14 ■ Gait 547

PERONEUS LONGUS N=19

500 VASTUS LATERALIS N=15 400

400 300 MEAN(ᒒV)=56

300
MEAN (ᒒV)=61.5 200
200
100
100
0 0

20

40

60

80
0

0
100

100
20

40

60

80
400 VASTUS MEDIALIS N=24 400 PERONEUS BREVIS N=10

300 300
MEAN (ᒒV)=117.3
MEAN (ᒒV)=50.4
200 200

100 100

0 0

100

0
20

40

60

80

100
0

20

40

60

80
400 RECTUS FEMORIS N=28 400 MEDIAL GASTROCNEMIUS N=25
MEAN (ᒒV)=110.7
300 MEAN (ᒒV)=21.8 300

200 200

100 100

0 0

100
100

20

40

60

80
20

40

60

80

0
0

TIBIALIS ANTERIOR N=26 LATERAL GASTROCNEMIUS

350 400 N=10
300
250 MEAN (ᒒV)=139 300 MEAN (ᒒV)=79.2
200
200
150
100 100
50
0 0 100
20

40

60

80
0
20

40

60

80
0

100

Figure 14-23 Electromyographic activ- EXTENSOR DIGITORUM LONGUS SOLEUS N=18

300 400
ity profiles of major muscle groups of the N=12
250 MEAN (ᒒV)=113
knee and ankle during one stride of gait 300
with probable roles. Signals were derived 200 MEAN (ᒒV)=98
from bipolar surface electrodes as data 150 200
normalized to means for each subject. 100
(Redrawn from Winter DA: The Biomechanics 100
50
and Motor Control of Human Gait: Normal,
0 0
Elderly and Pathological [ed. 2]. Waterloo,
Ontario, Waterloo Biomechanics, 1991, with
80
60
40
0

20

100
0
100
20

40

60

80

permission from David A. Winter.)

2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 548

548 SECTION 5 ■ Integrated Function

peaking around 70% of the cycle, reflecting this hip-flexor related to the moment profiles appearing in Figures 14–16
function. and 14–18. The GRFVs, joint positions, and muscle activ-
Now let us look at the major energy-absorbing phases ity that were used to create the illustrations were derived
and the muscle groups that are responsible. K1-S, occurring from published studies on normal human walking.4,16
before 20% of the cycle, is the eccentric phase of the knee
extensors (vastus lateralis, medialis, intermedius, and
rectus femoris), which precedes its concentric K2-S en-
ergy-generating phase. We have discussed H2-S, the eccen- CASE APPLICATION
tric action of the hip flexors (iliopsoas and rectus Hip and Knee
femoris), during midstance and late stance. Note that the
knee extensor and the hip extensor muscles begin their
contraction at the end of swing phase, although at this time
the dominant moments are being provided by the knee flex-
Flexors
case 14–6
At heel-off (~40% of gait cycle), the GRFV tending to extend
ors. K3-S is a small energy-absorbing internal knee extensor the hip and knee (see Fig. 14–12) is consistent with the
moment occurring while the knee is flexing rapidly (~50% opposition provided by the hip flexor moment (iliopsoas
to 70% of cycle), and this is reflected in low levels of con- and rectus femoris muscles) and the knee flexor moment
traction of the vastus muscles, particularly the rectus (gastrocnemius muscles) (see Fig. 14–16). Consider the
femoris, which, because it crosses both hip and knee, is effects of a larger than normal tendency to extend the knee,
active at that time as a hip flexor (H3-S). K4-S, energy ab- which occurred, for example, when Ms. Brown thrust her
sorption of the knee flexors at the end of swing, is reflected knee back into full extension in an effort to gain knee stabil-
in EMG records as activation of the medial and lateral ity (Fig. 14–25). Because the knee flexors that are active at
hamstrings. Note that the gastrocnemius muscles, which that time (gastrocnemius muscles) did not overcome this
cross the knee as well as the ankle, also begin activity in late excessive moment, Ms. Brown had difficulty flexing the
swing phase. At the ankle, A1-S absorption through much of knee, which prevented flexion of both the hip and the ankle
early stance and midstance is attributable to ankle plan- and reduced the opportunity to generate work from the an-
tarflexor activity, which we have already discussed. kle at A2-S and the hip at H3-S. Avoiding this “knee locking”
Now let us see whether we have missed any major EMG in stance by rigorous encouragement of knee flexion at
features shown in Figures 14–22 and 14–23 by deducing the end of stance and by temporary use of an ankle-foot
muscle activity from what we know about support moment orthosis was important in Ms. Brown’s gait reeducation.
and power profiles. First, in the sagittal plane, we have
not considered the ankle dorsiflexors (tibialis anterior,
extensor digitorum longus, and extensor hallucis longus Concept Cornerstone 14-3
muscles), which are eccentrically active in early stance be-
fore foot flat and act to lower the foot to the floor. They are What Gait Information Is Important?
Given the availability of information about walking, it is
often not clear what information is helpful for any given
situation. First, it is important to determine why you want
Continuing Exploration 14-6:
the information. Usually the reasons include one or more
Ground Reaction Force and of the following: (1) to gain an understanding of normal
or pathological gait; (2) to assist movement diagnosis and
Muscle Activity identify specific causes of pathological gait; (3) to inform
Before full dynamic biomechanical gait analyses were treatment selection; and (4) to evaluate the effectiveness
common, attempts were made to link the static informa- of treatment. Second, you may ask what gait measures are
tion from the GRFV to the joint positions during gait important for the situation. For example, if you want to
in a visual way. There are errors in this kind of analysis know whether energy costs are decreased with provision
because dynamic factors are not included, but during of an ankle orthosis, a measure of self-selected speed of
stance phase, these are minimal. Also, errors caused by walking may be sufficient. If you want to know which
using the GRFV are less at joints that are nearer the force muscle groups could be exploited to gain increased walk-
platform. Errors are small at the ankle but become larger ing speed in treating a person with a neurological condi-
at the hip, especially at times of push-off and initial tion, you would want to see a power analysis. If you
contact.55 If they are used to attempt to reconcile with the wished to know whether a new ankle-foot orthosis really
internal moment profiles of Figures 14–11, 14–12, 14–16, did return energy during push-off, you would also want a
and 14–18, however, they can add important understand- power analysis. If you wanted to determine whether sur-
ing and helpful visualization of normal gait for stance gery to realign the tibia and fibula was successful in de-
phase of gait (of course, there is no GRF during swing creasing a varus or valgus moment on the knee, you would
phase). The general sequence of the most common pat- want a moment analysis in the frontal plane. There is a
tern of GRFV in the sagittal plane for stance phase is tendency for power analyses to be more useful in neuro-
shown in Figure 14–11 (initial contact to midstance) and logical conditions and moment analyses in musculoskele-
Figure 14–12 (midstance to toe-off). These should be tal conditions, particularly those involving pain.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 549

CHAPTER 14 ■ Gait 549

active again during swing phase to hold the foot at a neutral We have so far not noticed adductor longus and bre-
angle, and show varying but small levels of activity at other vis muscles, which act in both the frontal and sagittal
times of the cycle, probably positioning the foot. We have planes, and show two fairly equal peaks of activity at ~10%
not so far noted the activity in the frontal plane of the hip and 65% to 80% of the gait cycle. The first peak is concur-
abductor gluteus medius, along with gluteus minimus rent with the hip abductors and may be providing stabi-
and tensor fasciae latae muscles (not shown). These mus- lization of the hip joint; the second occurs early in swing,
cles control the lateral drop of the pelvis on the side of the providing hip flexion to assist iliopsoas and rectus
swinging leg. Activity of these muscles diminishes during femoris muscles. Further information can be found in the
midstance and ceases when the opposite limb has contacted literature.37,39,40 Trunk muscle activity is discussed in a
the ground.14 later section.

Table 14–3 Summary of Gait Characteristics From Initial Contact to Midstance

SAGITTAL PLANE ANALYSIS
Joint Motion Ground Reaction Internal Power Major Muscle
Force Moment Activity
Hip Extends from Anterior to posterior Extensor to neutral Generation (hip Gluteus
+20°–+30° to 0° extensors) maximus
Hamstrings
Knee Flexes from 0° to Anterior to posterior Flexor, then extensor Generation (knee flexors) Hamstrings
+15°, then to anterior then flexor Absorption (knee Quadriceps
extends +15° extensors) Hamstrings
to +5° Generation (knee Gastrocnemius
extensors)
Absorption (knee flexors)
Ankle Plantarflexes from Posterior to anterior Dorsiflexor, then Absorption (dorsiflexors) Tibialis anterior
0° to –5°, then plantarflexor Generation Soleus
dorsiflexes to +5° (plantarflexors) Gastrocnemius

FRONTAL PLANE ANALYSIS

Joint Motion Ground Reaction Moment Power Major Muscle
Force Activity
Hip Adduction from Medial Abductor Absorption (hip Gluteus medius
neutral to +5°, abductors)
then abduction Generation (hip
to 3° abductors)
Knee Small variation Medial Abductor Small generation Tensor fasciae
around neutral (knee abductors) latae
Ankle Inconsistent

TRANSVERSE PLANE ANALYSIS

Joint Motion Moment Power Major Muscle
Activity
Hip Externally rotates a External rotator Absorption (internal Tensor fasciae latae
few degrees rotators) (Ligamentous)
Gluteus medius
Gluteus minimus
Knee From several Very small external Small absorption
degrees of rotator (external rotators)
external rotation,
rotates internally
to neutral
Ankle Inconsistent

Redrawn from Eng JJ, Winter DA: Kinetic analysis of the lower limbs during walking: What information can be gained from a three dimensional model? J Biomech 28:753, 1995;
and from Winter DA, Eng JJ, Isshac MG: A review of kinetic parameters in human walking. In Craik RL, Otis CA (eds): Gait Analysis: Theory and Application. St. Louis, MO,
Mosby–Year Book, 1994.
Values are for young males.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 550

550 SECTION 5 ■ Integrated Function

Table 14–4 Summary of Gait Characteristics From Midstance to Toe-Off

SAGITTAL PLANE ANALYSIS
Joint Motion Ground Reaction Moment Power Major Muscle
Force Activity
Hip Extends from 0° Posterior Flexor Absorption (flexors) Iliopsoas
to –20° Generation (flexors) Rectus femoris
extension, then Iliopsoas
begins to flex Rectus femoris
Knee From +5° extends Posterior Flexor, then extensor Absorption (extensors) Vastus
a few degrees, Small generation (knee Gastrocnemius
then flexes to flexors) Rectus femoris
about +45° Large absorption
(extensors)
Ankle From +5° Anterior Plantarflexor Very small absorption Soleus
dorsiflexion, (plantarflexors) Gastrocnemius
dorsiflexes a few Large generation Soleus
more degrees, (plantarflexors) Gastrocnemius
then rapidly plan-
tarflexes to –25°

FRONTAL PLANE ANALYSIS

Joint Motion Ground Reaction Moment Power Major Muscle
Force Activity
Hip From +3° smoothly Medial Abductor Generation (abductors) Gluteus
abducts to –5° medius
Knee Neutral, then Medial Abductor Absorption (abductors) Tensor fasciae
abducts to a few latae,
degrees –3° ligaments
Ankle Inconsistent

TRANSVERSE PLANE ANALYSIS

Joint Motion Moment Power
Hip From –3°, external Internal rotator Very small
rotation a few absorption
more degrees (external rotators)
Knee Remains near Internal rotator Very small
neutral absorption
(ligamentous)
Ankle Inconsistent

Redrawn from Eng JJ, Winter DA: Kinetic analysis of the lower limbs during walking: What information can be gained from a three dimensional model? J Biomech 28:753, 1995;
and from Winter DA, Eng JJ, Isshac MG: A review of kinetic parameters in human walking. In Craik RL, Otis CA (eds): Gait Analysis: Theory and Application. St. Louis, MO,
Mosby–Year Book, 1994.
Values are for young males.

Gait Initiation and Termination initiation of movement to the beginning of the gait cycle.
Gait initiation begins in the erect standing posture with an
The previous sections have discussed gait as a continuing ac- inhibition of the gastrocnemius and soleus muscles closely
tivity. However, both starting to walk and stopping walking followed by activation of the tibialis anterior muscles.56
require a certain sequence of motor events. These events are Bilateral concentric contractions of the tibialis anterior
interesting because they are quite different from continuing muscles (pulling the tibias forward over stable feet) results
gait, they are potentially destabilizing to a person’s balance, in a sagittal moment that inclines the body anteriorly from
and yet they are essential to the function of walking. the ankles. Initially, the CoP shifts posteriorly, and briefly,
Gait initiation is defined as a stereotyped activity that in- toward the first swing foot from its resting location between
cludes the series or sequence of events that occurs from the the feet and anterior to the ankle joints. As the heel of the
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 551

CHAPTER 14 ■ Gait 551

180 SPLENIUS CAPITUS N=12 250 TRAPEZIUS (UPPER) N=11

160 MEAN (ᒒV)=24.2
140 200
120
100 150
80 100
60
40 MEAN (ᒒV)=13.9 50
20
0 0

20

40

60

80
20

40

60

80

100
0

0
100
ERECTOR SPINAE-T9 ERECTOR SPINAE-L3-L4
500 300
N=12 N=11
400 MEAN (ᒒV)=39 MEAN (ᒒV)=27.7
200
300
200
100
100
0 0
0

20

40

60

80
100

0
20

40

60

80

100
250 EXTERNAL OBLIQUE- LATERAL 180 RECTUS ABDOMINUS N=11
Figure 14-24 Electromyographic activity N=10 160
profiles of major muscle groups of the 200 140
trunk during one stride of gait. Signals were 120
150 100
derived from bipolar surface electrodes as
80
data normalized to means for each subject. 100
60
(Redrawn from Winter DA: The Biomechanics 50 40 MEAN (ᒒV)=7.3
and Motor Control of Human Gait: Normal, MEAN (ᒒV)=24.3 20
Elderly and Pathological [ed. 2]. Waterloo, 0 0
Ontario, Waterloo Biomechanics, 1991, with
100

20

40

60

80
20

40

60

80

0
0

100
permission from David A. Winter.)

swing foot lifts off the floor the CoP moves rapidly to the vastus lateralis is usually activated first.7 Braking forces are
stance foot heel; then, as the swing foot toe lifts off the floor, then produced by the swing limb at initial contact with a
it progresses anteriorly to the forefoot.6 Abduction of the large soleus muscle burst to control ankle plantarflexion,
swing hip, with activation of the gluteus medius, occurs al- and reduced activity in the tibialis anterior to inhibit ankle
most simultaneously with contractions of the tibialis ante- dorsiflexion. To maintain knee extension and stabilize the
rior. As weight is transferred to the stance limb, activiaton trunk the vastus lateralis activity is continued and the erec-
of gluteus medius on the stance side begins. According to tor spinae are activated.
Elble and colleagues,57 the support limb hip and knee flex a
few degrees (3° to 10°), and the CoP moves anteriorly and
medially toward the support limb. A healthy individual may CASE APPLICATION
initiate gait with either the right or left lower extremity, and Avoiding Instability
no changes will be seen in the pattern of events. The tem- in Initiation
poral and spatial patterns of gait initiation are preserved in
the elderly, though there is a trend for displacements to be
smaller and velocities lower.6
of Gait
case 14–7
Patients with hemiparesis, like Ms. Brown, demonstrate
Termination of gait has received much less attention than some differences in gait initiation depending on whether the
has initiation (for a review, see Sparrow and Tirosh, 2004).7 step is started with the affected or unaffected limb. When
Anticipation of stopping appears to influence EMG activity the person stands on the unaffected limb and steps forward
even before initial contact.58 The body attempts to maintain with the affected leg, the timing and pattern of events
the stance limb anterior to the whole body center of are practically the same as they would be in a nonaffected
mass, which is initiated first in the stance limb by reducing person, but when the person with paresis attempts gait
push-off and activating hip extensors and knee extensors initiation with the nonaffected leg, the weight must first
as energy absorbers. The reduced push-off in the stance be shifted onto the affected side. The result is an erratic pat-
limb is accomplished by inhibition of soleus and strong tern of events, and stability can be seriously threatened.59
activation of the tibialis anterior.58 In the swing limb, the
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 552

552 SECTION 5 ■ Integrated Function

during swing phase of the right limb, goes through neutral

position about midstance, then continues rotating the same
amount in preparation for left foot initial contact. When the
pelvis rotates counterclockwise, preparing for initial contact
of the right foot, the trunk rotates clockwise with regard to
the pelvis to keep it directed forward. The amount of trans-
verse rotation of the trunk during gait is slight and occurs
primarily in a direction opposite to the direction of pelvic
rotation (Fig. 14–26). As the pelvis rotates forward with the
swinging lower extremity, the thorax on the opposite side
rotates forward as well. Actually, the thorax undergoes a
biphasic rotation pattern with a reversal directly after liftoff
of the stance leg. The thorax is rotated backward during
double support and then slowly rotates forward during sin-
gle support. This trunk motion helps prevent excess body
motion and helps counterbalance rotation of the pelvis.
Krebs and coworkers found that at a free speed of gait,
transverse rotation reached a maximum of 9° at 10% of the
cycle after each heel strike.61 In a study of treadmill walking,
Stokes and associates62 found that the movements and inter-
actions of the trunk and pelvis were extremely complex
when translatory and rotatory movements of the trunk were
CoP
considered along with anterior and posterior pelvic tilting,
Figure 14-25 GRFV passing excessively anterior to knee joint lateral pelvic tilting, and rotation. Mediolateral translations
center, making it difficult for the person to flex the knee. of the trunk occur as side-to-side motions (leans) in relation
to the pelvis. For example, the trunk is leaning or moving to

TRUNK AND UPPER EXTREMITIES

Trunk leans right
Trunk 2-3 cm

The trunk remains essentially in the erect position during

normal free-speed walking on level ground, varying only
1.5°.60 Krebs and coworkers61 found that a flexion peak of
low amplitude occurred near each heel strike and an exten-
sion peak of low amplitude occurred during single limb sup-
port. Winter explained17 that at initial contact, the forward Trunk rotates
acceleration of the head, arms, and trunk is quite large and a few degrees
acts at a distance from the hip joint, thus producing an un-
balancing moment that acts strongly to cause flexion of the
trunk during weight acceptance. However, an almost equal Right arm swings
back about 24˚
and opposite moment is provided by the hip extensors
Left arm swings
(which we have seen before in their role in support and in forward about 6˚
energy generation at H1-S). In the frontal plane, a similar Pelvis rotates
situation exists. The center of mass of the head, arms, and 4˚- 8˚
trunk, always being medial to the hip joint, exerts a consid-
erable moment that is balanced primarily by an internal hip
abductor moment from the supporting limb (see Fig. 14–18)
and is assisted by the medial acceleration of the hip.
The biomechanical models used previously in the chap-
ter do not distinguish between the pelvis and the trunk. In
the sagittal plane, the pelvis moves in a sinusoidal curve up Right Left
and down 4 or 5 cm with each step, the low point coincid-
ing with initial contact of each foot and the high point coin-
ciding with midstance. In the frontal plane, the pelvis trans-
lates from side to side about the same amount toward the
standing limb and rotates downward about 5° toward the
swinging limb on each side. In the transverse plane, looking Figure 14-26 Pelvis, trunk, and arm motion. Note that the
at the pelvis from above, it rotates 4° to 8° counterclockwise trunk and arms rotate in a direction opposite to that of the pelvis.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 553

CHAPTER 14 ■ Gait 553

the right from right heel strike to left toe-off, at which point rotators are acting eccentrically to control external rota-
the trunk begins a lean to the left until right toe-off. The tion of the arm at the shoulder as the posterior deltoid
average total ROM that occurs during the mediolateral acts eccentrically to restrain the forward swing. The latis-
trunk leans is about 5.4 cm.62 Hirasaki and coworkers63 used simus dorsi and teres major muscles, as well as the
a treadmill and a video-based motion analysis system to posterior deltoid, may then act concentrically to produce
study trunk and head movements at different walking the backward swing. The role of the middle deltoid is
speeds. These authors found that the relationship between unclear, although it has been suggested that it functions to
walking speed and head and trunk movements was the most keep the arm abducted so that it may clear the side of the
linear in the range of walking speeds from 1.2 to 1.8 m/sec. body.45 Activity in all muscles increases as the speed of gait
At velocities above and below this range, head and trunk increases.65
movements were less well coordinated. Recent work supports neuronal coordination of arm and
EMG profiles and probable functional roles are shown leg movement during locomotion. Deitz and colleagues68
for some trunk muscles in Figure 14–24. Recent EMG and Wannier and colleagues69 reported the behaviors of the
studies on the trunk muscles during gait have shown that arm to leg corresponding to a system of two coupled oscil-
there are subgroups of subjects who show similar patterns lators. Results were compatible with the assumption that the
of muscle activity64 when cluster analysis is used on the proximal arm muscles are associated with the swinging of
EMG data. This applies to the internal oblique, external the arms during gait as a residual function of quadripedal
oblique, rectus abdominis, and lumbar erector spinae locomotion.
muscles. Although most subjects showed low levels of ac-
tivity throughout the gait cycle, the internal oblique and TREADMILL, STAIR, AND RUNNING
erector spinae muscles had more distinct bursts, usually
occurring close to initial contact. Some researchers have GAITS
shown two periods of activity65 for the erector spinae
muscle, one at initial contact and the second at toe-off. It Treadmill Gait
is thought that its function is to oppose the unbalancing
moment that acts strongly to cause flexion of the trunk dur- Treadmills used for gait measurement and training have a
ing weight acceptance. number of advantages including a small footprint, the avail-
ability of weight support for patient safety, and increasing
sophistication of instrumentation that may include meta-
Upper Extremities bolic analysis as well as embedded force platforms. Compar-
Detailed kinetics of the upper extremity during normal isons of treadmill with overground walking have shown
gait have not been reported, although extensive mapping higher cadence and shorter stance times at comparable
of EMG was classically performed several decades ago.65,66 speeds,70,71 although others do not report these differ-
Although the lower extremities are moving alternately ences.72,73 Differences in joint ranges of motion have also
forward and backward, the arms are swinging rhythmically. been reported, but these have usually been less than 2.5° and
However, the arm swinging is opposite to that of the legs not considered important.70,74 Kinetic comparisons have re-
and pelvis but similar to that of the trunk (see Fig. 14–26). vealed similar joint moments for treadmill and overground
The right arm swings forward with the forward swing of the walking, although push-off forces and GRF maxima have
left lower extremity while the left arm swings backward. generally been lower with treadmill walking.73,74,75 It is still
This swinging of the arms provides a counterbalancing not certain whether metabolic costs are different. Whereas
action to the forward swinging of the leg and helps to decel- Pearce and colleagues75 reported lower oxygen uptake in as-
erate rotation of the body, which is imparted to it by the sociation with treadmill walking compared to overground
rotating pelvis. The total ROM at the shoulder is not very walking at the same speed in young and older men, Parvata-
large. At normal free velocities, the ROM is only approxi- neni and colleagues76 found significantly higher metabolic
mately 30° (24° of extension and 6° of flexion). It is clear, costs on the treadmill in healthy older adults when speeds
however, that resticting arm motion increases energy costs on the treadmill were matched with overground speeds.76
of walking at prescribed speeds.67 These findings are consistent with reports of higher heart
The normal shoulder motion is the result of the com- rates by Grieg and colleagues.72 More research is needed to
bined effects of gravity and muscle activity. During the assess energetic differences between treadmill and over-
forward portion of arm swinging, the following medial ground walking across age groups, and to determine their
rotators are active: subscapularis, teres major, and latis- cause.
simus dorsi muscles. In backward swing, the middle and Stair Gait
posterior deltoid muscles are active throughout, and the
latissimus dorsi and teres major muscles are active only Ascending and descending stairs is a basic body movement
during the first portion of backward swing.65 The required for performing normal activities of daily living
supraspinatus, trapezius,65 and posterior and middle such as shopping, using public transportation, or simply get-
deltoid muscles65 are active in both backward and ting around in a multistory home or building. Although
forward swing. It is interesting to note that little or no many similarities exist between level-ground locomotion
activity was reported in the shoulder flexors in these and stair locomotion, the difference between the two modes
studies.65,66 It appears that during forward swing, the medial of locomotion may be significant for a patient population.
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 554

554 SECTION 5 ■ Integrated Function

The fact that a patient has adequate muscle strength and into two subphases.8 The subdivisions of the stance phase
joint ROM for level-ground walking does not ensure that are weight acceptance, pull-up, and forward continuance.
the patient will be able to walk up and down stairs. Stair The subdivisions of the swing phase are foot clearance
walking represents additional stress over level-ground walk- and foot placement. As can be seen in Figure 14–27,
ing and may reveal differences that are not apparent in level- weight acceptance comprises approximately the first 14%
ground walking.77 of the gait cycle and is somewhat comparable to the heel
Krebs and coworkers61 found that trunk ROM during strike throughout the loading phase of walking gait. How-
level-ground gait was similar to trunk ROM during stair de- ever, in contrast to walking gait, the point of initial con-
scent but differed from trunk ROM during stair ascent in all tact of the foot on the stairs is usually located on the an-
planes. The maximum ROM of trunk flexion in relation to terior portion of the foot and travels posteriorly to the
the room during stair ascent was at least double the amount middle of the foot as the weight of the body is accepted.
of trunk flexion found in either stair descent or in level- The pull-up portion, which extends from approximately
ground walking.61 14% to 32% of the gait cycle, is a period of single-limb
Locomotion on stairs is similar to level-ground walk- support. The initial portion of pull-up is a time of insta-
ing in that stair gait involves both swing and stance phases bility, inasmuch as all of the body weight is shifted onto
in which forward progression of the body is brought the stance extremity when it is flexed at the hip, knee, and
about by alternating movements of the lower extremities. ankle. During this period, the task is to pull the weight of
Also, in both stair and level-ground gait, the lower the body up to the next stair level. The knee extensors are
extremities must balance and carry along the head, arms, responsible for most of the energy generation required to
and trunk. McFayden and Winter8 (using step dimensions accomplish pull-up. The forward continuance period is
of 22 cm for the stair riser and 28 cm for the tread) from approximately 32% to 64% of the gait cycle and
performed a sagittal plane analysis of stair gait.8 These corresponds roughly to the midstance through toe-off
investigators collected kinetic and kinematic data for one subdivisions of walking gait. In the forward continuance
subject during eight trials. The stair gait cycle for stair period, the greatest amount of energy is generated by
ascent presented in Figure 14–27 is based on data from the ankle plantarflexors. Data regarding joint ROM
McFayden and Winter’s study.8 Although ankle moments and muscle activity for ascending stairs are presented in
were aproximately the same, the knee extensor moment in Tables 14–5, 14–6, and 14–7.8 Similar findings have been
both stair ascent and descent was approximately three reported by Protopapadaki,78 Costigan,79 Salsich,80 and
times larger than that of level-ground walking. This may their colleagues. Variations between studies have been
cause particular mobility problems in patients with knee attributed to differences in stair dimensions, trunk incli-
pain, which is common in such conditions as osteoarthritis nation, foot contact positions, and differences in subject
and patello-femoral pain syndrome. Although the knee characteristics and methodologies.
has primarily an absorptive function in level walking, it
has a large generative role in stair ascent. Powers are
largely generative in stair ascent and absorptive in descent Concept Cornerstone 14-4
for all joints.
The investigators divided the stance phase of the stair
Differences Between Level-Ground
gait cycle into the three subphases and the swing phase Gait and Stair Gait
Table 14–6 shows that considerably more hip and knee
flexion are required in the initial portion of stair gait
than are required in normal level-ground walking.
Therefore, a patient would require a greater ROM
for stair climbing (the same stair dimensions and slope)
than for normal level-ground walking. Naturally muscle
activity and joint ROMs will change if stairs of other
dimensions are used.8

Ascending stairs involves a large amount of positive work

that is accomplished mainly through concentric action of
the rectus femoris, vastus lateralis, soleus, and medial gas-
trocnemius muscles. Descending stairs is achieved mostly
through eccentric activity of the same muscles and involves
energy absorption. The support moments during stair as-
cent, stair descent, and level-ground walking exhibit similar
patterns; however, the magnitude of the moments is greater
in stair gait, and, consequently, more muscle strength is re-
Figure 14-27 Stair gait cycle. quired. Kirkwood and colleagues81 found that the maximum
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 555

CHAPTER 14 ■ Gait 555

Table 14–5 Sagittal Plane Analysis of Stair Ascent (Fig. 14–29): Stance Phase–Weight
Acceptance (0%–14% of Stance Phase) Through Pull-Up (14%–32% of
Stance Phase)
JOINT MOTION MUSCLE CONTRACTION
Hip Extension: 60°–30° of flexion Gluteus maximus Concentric
Semitendinosus
Gluteus medius
Knee Extension: 80°–35° of flexion Vastus lateralis Concentric
Rectus femoris
Ankle Dorsiflexion: 20°–25° of dorsiflexion Tibialis anterior Concentric
Plantarflexion: 25°–15° of dorsiflexion Soleus Concentric
Gastrocnemius

Table 14–6 Sagittal Plane Analysis of Stair Ascent (Fig. 14–29): Stance Phase–Pull-Up
(End of Pull-Up) Through Forward Continuance (32%–64% of the Stance
Phase of Gait Cycle)
JOINT MOTION MUSCLE CONTRACTION
Hip Extension: 30°–5° flexion Gluteus maximus Concentric and isometric
Flexion: 5° to 10°–20° of flexion Gluteus medius Eccentric
Semitendinosus
Gluteus maximus
Gluteus medius
Knee Extension: 35°–10° of flexion Vastus lateralis Concentric
Flexion: 5° to 10°–20° of flexion Rectus femoris Eccentric
Rectus femoris
Vastus lateralis
Ankle Plantarflexion: 15° of dorsiflexion to Soleus Concentric
15°–10° of plantarflexion Gastrocnemius Eccentric
Tibialis anterior

Table 14–7 Sagittal Plane Analysis of Stair Ascent (Fig. 14–29): Swing Phase
(64%–100% of Gait Cycle)—Foot Clearance Through Foot Placement
JOINT MOTION MUSCLE CONTRACTION
Hip Flexion: 10°–20° to 40°–60° of flexion Gluteus medius Concentric
Extension: 40°–60° of flexion to
50° of flexion
Knee Flexion: 10° of flexion to 90°–100° Semitendinosus Concentric
of flexion Vastus lateralis Concentric
Extension: 90°–100° of flexion to 85° Rectus femoris
of flexion
Ankle Dorsiflexion: 10° of plantarflexion to Tibialis anterior Concentric and
20° of dorsiflexion isometric

peak internal abductor moment at the hip occurred during the case of stair gait, a patient who is able to walk on level
descending stairs. ground may not have the ability to run. Running requires
greater balance, muscle strength, and ROM than does
normal walking. Greater balance is required because run-
Running Gait ning is characterized not only by a considerably reduced
Running is another locomotor activity that is similar to base of support but also by an absence of the double-support
walking, but certain differences need to be examined. As in periods observed in normal walking and the presence of
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 556

556 SECTION 5 ■ Integrated Function

float periods in which both feet are out of contact with the moments, and powers between walking and running are
supporting surface (Fig. 14–28). The walking gait cycle in shown in Figure 14–29.85 At the beginning of the stance
Figure 14–2 can be used to compare the gait cycle in walk- phase of running, the hip is in about 45° of flexion at heel
ing with running gait. The percentage of the gait cycle spent strike and extends during the remainder of the stance phase
in float periods will increase as the speed of running in- until it reaches about 20° of hyperextension just after toe-
creases. Muscles must generate greater energy both to raise off.84 The hip then flexes to reach about 55° to 60° of flex-
the head, arms, and trunk higher than in normal walking ion in late swing. Just before the end of the swing phase, the
and to balance and support them during the gait cycle. Mus- hip extends slightly to 45° to 50° in preparation for heel
cles and joint structures also must be able to absorb more strike.62 The knee is flexed to about 20° to 40° at heel strike
energy to accept and control the weight of the head, arms, and continues to flex to 60° during the loading response.
and trunk. Thereafter, the knee begins to extend, reaching 40° of flex-
For example, in normal level-ground walking, the mag- ion just before toe-off. During the swing phase and initial
nitudes of the vertical GRFs at the CoP at initial contact are float period, the knee flexes to reach a maximum of approx-
approximately 70% to 80% of body weight and rarely ex- imately 125° to 130° in the middle of the swing phase. In
ceed 120% of body weight during the gait cycle.82,83 How- late swing, the knee extends to 40° in preparation for heel
ever, during running, the GRFs at the CoP have been strike.83 In Figure 14–29, note these differences in joint an-
shown to reach 200% of body weight and increase to 250% gles from level-ground walking. At each joint, the maxima
of body weight during the running cycle. Furthermore, the for running exceed those of walking. In the case of the hip,
knee is flexed at about 20° when the foot strikes the ground. only flexion range is increased. The knee range in running
This degree of flexion helps to attenuate impact forces but is not very different from that in walking, but it takes place
also increases the forces acting at the patellofemoral joint. in approximately 25° more flexion. The ankle has greatly in-
In addition, the base of support in running is considerably creased dorsiflexion and modestly increased plantarflexion.
less than in walking. A typical base of support in walking is The ankle is in about 10° of dorsiflexion at heel strike and
about 2 to 4 inches, whereas in running, both feet fall in the rapidly dorsiflexes to reach about 25° to 30° dorsiflexion.
same line of progression, and so the entire center of mass of The rapid dorsiflexion is followed immediately by plan-
the body must be placed over a single support foot. To com- tarflexion, which continues throughout the remainder of the
pensate for the reduced base of support, the functional limb stance phase and into the initial part of the swing phase.
varus angle increases. Functional limb varus angle is the an- Plantarflexion reaches a maximum of 25° in the first few sec-
gle between the bisection of the lower leg and the floor.89 onds of the swing phase. Throughout the rest of the swing
According to McPoil and Cornwall,84 the functional limb phase, the ankle dorsiflexes to reach about 10° in late swing
varus angle increases about 5° during running in compari- in preparation for heel strike.83
son with walking. The whole extremity begins to medially rotate during the
swing phase. At heel strike, the extremity continues to me-
Joint Motion and Muscle Activity dially rotate and the foot pronates. Lateral rotation of the
stance extremity and supination of the foot begins as the
Joint Motion swing leg passes the stance limb in midstance. The ROM in
The ROM varies according to the speed of running and the lower extremities needed for running, in comparison
among different researchers. Comparisons in joint angles, with the ROM required for normal walking, is presented in
Table 14–8. The largest differences in the total ROM re-
quirements between the two activities appear to be at the
knee and hip joints. At the knee joint, up to an additional
RUNNING CYCLE PHASES 90° of flexion is required for running versus walking. At the
hip joint, running requires about twice the amount of mo-
tion that was needed for normal walking.

Muscle Activity
The gluteus maximus and gluteus medius muscles are ac-
tive both at the beginning of the stance phase and at the end
of the swing phase. The tensor fasciae latae muscle is also
active at the beginning of stance and at the end of swing but
also is active between early and midswing. The adductor
magnus muscle shows activity for about 25% of the gait cy-
cle from late stance through the early part of the swing
phase. Activity in the iliopsoas muscle occurs for about the
same percentage of the gait cycle as the adductor longus
muscle, but iliopsoas activity also occurs during the swing
phase from about 35% to 60% of the gait cycle.
The quadriceps muscle acts eccentrically during the first
10% of the stance phase to control knee flexion when the
Figure 14-28 Running gait cycle. knee is flexing rapidly. The quadriceps ceases activity after
 2362_Ch14-524-568.qxd
Sagittal Sagittal Sagittal
Joint Angle Joint Moment Joint Power

15
85

flexion
2

gen
10

ext
65

DEGREES
1 5

2/2/11
Watt/kg
45
HIP

N.m/kg
0 0
25 -1 -5

extension

4:44 PM
5 -10

abs
-2

flex
-15 -3 -15
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

Page 557
110 15
flexion

ext

gen
10
DEGREES

80 1 5
KNEE

Watt/kg
N.m/kg
0 0
50
-1 -5
extension

20

abs
-2

flex
-10

-10 -3 -15
0 20 40 60 80 100 0 20 40 60 80 100 0 20 40 60 80 100

40 3 15
dorsiflexion

gen
2
ext

10
20 1 5
DEGREES

Watt/kg
N.m/kg
ANKLE

0 0 0
-1 -5
20
plantar

abs
flex

-2 -10

-40 -3 -15
0 20 40 60 80 0 20 40 60 80 0 20 40 60 80

% Gait Cycle
% Gait Cycle % Gait Cycle

Figure 14-29 Sagittal plane joint angles, moments, and powers for running (solid line) and walking (dotted line).
Joint angles are plotted with flexion and dorsiflexion positive. Moments are internal moments normalized to body mass, plotted with extensor and plan-
tarflexor positive. Powers are normalized to body mass, plotted with generation positive and absorption negative. The vertical solid line near 40% of gait
cycle represents toe-off for running; the vertical dotted line near 60% of gait cycle represents toe-off for walking. (Joint angles, moments, and powers for
557

running redrawn from Novacheck TF: The biomechanics of running. Gait Posture 7:77, 1998, with permission from Elsevier.)
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 558

558 SECTION 5 ■ Integrated Function

tibialis anterior activity between walking and running gaits.

Table 14–8 Average Peak ROM at the
Tibialis anterior activity in the first half of the stance phase
Hip, Knee, and Ankle: in running gait accounts for the remainder of the difference
Comparison Between in activity in this muscle.
Running69–71 and Walking8,71
RUNNING WALKING
Moments, Powers, and Energies
Hip joint Hip joint In Figure 14–29, note the differences in internal moments
Flexion 55°–65° Flexion 30° and powers for running from those of walking (see
Extension 10°–20° Extension 0°–20° Figs. 14–16 and 14–19). In all joints, the moments are
Knee joint greatly increased, particularly the knee extensor moment in
Flexion 80°–130° Knee joint early stance. Similarly, all the power generators are greatly
Extension 0°–5° Flexion 40°–50° increased in running. The chief generators are, again, A2,
Extension 0° H3, and H1, but an additional source, K2, is now important.
Ankle joint Note that the knee extensors in early stance (K2) are impor-
Dorsiflexion 10°–30° Ankle joint tant in running. The absorption phases typical of walking
Plantarflexion 20°–30° Dorsiflexion 10° are increased at all joints in running.
Plantarflexion 20° In walking, potential and kinetic energy of the head,
arms, and trunk are out of phase, which results in consider-
able energy savings. In running, these savings cannot occur,
but two other methods of gaining efficiency are present.
First, there are storage and return of elastic potential energy
the first part of stance, and no activity occurs until the last by the stretch of elastic structures, particularly tendons.
20% of the swing phase, when concentric activity begins to Second, there is transfer of energy from one body segment
extend the knee (to 40° of flexion) in preparation for heel to another by two-joint muscles acting as “energy straps.”85
strike. The medial hamstrings are active at the beginning
of stance and through a large part of swing. For example,
the medial hamstrings are active from 18% to 28% of the
Summary
stance phase, from about 40% to 58% of initial swing, and As a result of the efforts of many investigators, our present
for the last 20% of swing. During part of this time, the knee body of knowledge regarding human locomotion is exten-
is flexing and the hip is extending, and the hamstrings may sive. However, gait is a complex subject, and further research
be acting to extend the hip and to control the knee. During is necessary to standardize methods of measuring and defin-
initial swing, the hamstrings are probably acting concentri- ing kinematic and kinetic variables, to develop inexpensive
cally at the knee to produce knee flexion, which reaches a and reliable methods of analyzing gait in the clinical setting,
maximum at midswing. In late swing, the hamstrings may be and to augment the limited amount of knowledge available
contracting eccentrically to control knee extension and to regarding kinematic and kinetic variables in the gaits of
reextend the hip. children and elderly persons.
A comparison of walking and running muscle activity at Standardization of equipment and methods used to quan-
the ankle shows that in walking, gastrocnemius muscle ac- tify gait variables, as well as standardization of the terms
tivity begins just after the loading response at about 15% of used to describe these variables, would help eliminate some
the gait cycle and is active to about 50% of the gait cycle of the present confusion in the literature and make it possi-
(just before toe-off). In running, gastrocnemius muscle ble to compare the findings of various researchers with some
activity begins at heel strike and continues through the first degree of accuracy. Some standardization has been provided
15% of the gait cycle, ending at the point at which activity by the manufacturers of motion analysis systems; as a result,
begins in walking. The gastrocnemius muscle becomes investigators using similar systems tend to use the same con-
active again during the last 15% of swing. ventions. At the present time, inexpensive, quantitative, and
The tibialis anterior muscle activity occurs in both reliable clinical methods of evaluating gait are limited to
stance and swing phases in walking and running. However, time and distance variables of step length, step duration,
the total period of activity of this muscle in walking (54% of stride length, cadence, and velocity.86 These measures pro-
the gait cycle) is less than it is in running, in which it shows vide a simple means for objective assessment of a patient’s
activity for about 73% of the gait cycle. The difference in status and for detecting overall change, but they provide
activity of the tibialis anterior muscle between walking and virtually no help in determining treatment. Increases in
running is due partly to the differences in the length of the step length and decreases in step duration may be used to
swing phases in the two types of gait. The swing period in document a patient’s progression toward a more normal gait
walking gait is approximately 40% of the total gait cycle, pattern; however, a normal gait pattern may not be appro-
whereas in running gait, the swing phase constitutes about priate for many patients. Although many movement practi-
62% of the total gait cycle. Most of the activity in the tib- tioners feel that it is best to aim for normal gait during early
ialis anterior muscle during both walking and running gait stages of rehabilitation, it is also important to identify the
is concentric or isometric action that is necessary to clear means by which a person can use the flexibility of the human
the foot in the swing phase of gait. The longer swing phase body to compensate for deficiencies.5,52 Automated gait
in running accounts for at least part of the difference in analysis computer programs can provide the clinician with
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 559

CHAPTER 14 ■ Gait 559

information about all of the kinematic and kinetic gait pa- factors in determining changes in time and distance meas-
rameters related to a particular subject or patient. However, ures with age. Average stride length was 76% of the child’s
the researcher or clinician must have sufficient knowledge height at a walking speed of 104 m/sec, regardless of a
of the kinematics and kinetics of normal gait to interpret child’s age. It is generally agreed that children’s gait patterns
and use the information for the benefit of the patient. have matured at least by the age of 7 years, with certain
variables considerably earlier.91,92 Studies involving young
children are difficult to perform and often complicated by
EFFECTS OF AGE, GENDER, ASSISTIVE the fact that the child’s musculoskeletal and nervous systems
DEVICES, AND ORTHOSES are in various stages of development. However, Sutherland
and colleagues attempted to provide evaluators with guide-
lines for assessing children’s gait by developing a group of
Age prediction regions for the kinematic motion curves in nor-
The gait of young children has received less attention than mal gait. A test of the prediction regions indicated that they
that of adults. The relatively few studies of children’s gait were capable of detecting a high percentage of abnormal
that have been conducted have shown that the age at which motion and therefore could be used as an initial screen to
independent ambulation begins is extremely variable among identify deficits in lower extremity function in children.93
individuals and that this variability continues throughout In contrast to the dearth of gait studies of young chil-
the developmental stages of walking. Cioni and coworkers87 dren, the effects of aging on gait continue to be the object
found that for 25 full-term infants, the age at which of many studies.20,45,91,95 Some of this interest in elderly gait
independent walking was attained (ability to move 10 has been prompted by the large number of hip fractures and
successive steps without support) varied between 12.6 and falls experienced by elderly persons. Fifty percent of elderly
16.6 months. In the first stage of independent walking, none people who were able to walk before a hip fracture are not
of the toddlers had heel strike, reciprocal arm swinging, or able to either walk or live independently after the fracture.96
trunk rotation. However, 4 months after attainment of inde- Furthermore, it is estimated that an elderly person experi-
pendent walking, nealy half of the children had heel strike, ences at least two falls per year,96 with the incidence higher
and nearly two thirds had reciprocal arm swinging and in women and the rate increasing with age.97 Increased sway
trunk rotation. and physiological decline are thought to be implicated and
The toddler has a higher relative center of mass than many falls occur during walking and while turning the
does the adult and walks with a wider base of support, a de- head.98 Therefore, many studies are directed toward deter-
creased single-leg support time, a shorter step length, a mining what constitutes normal elderly gait and whether
slower velocity, and a higher cadence in comparison with falls are caused by deficits in motor functioning or control,
normal adult gait. A study of 3- and 5-year-old children by other deficits that may accompany normal aging, or by
showed that some relationships between these variables adaptations employed that are thought to make their gait
were similar to the relationships found in adult gaits.88 For safer.99
example, as a group, the 3- and 5-year-old children showed The use of different age groups and levels of activity
significant increases in stride length adjusted for leg length, (sedentary versus active groups) among investigators has
step length, and cadence from a slow to a free speed and made it difficult to draw definitive conclusions about the
from a free to a fast speed of gait. In a study that included effects of normal aging. With respect to temporal and kine-
children from 6 to 13 years of age, Foley and associates89 re- matic variables, elderly persons, in comparison with younger
ported that the ROMs for flexion and extension of the joints groups, usually demonstrate a decrease in natural walking
of the lower extremities were almost identical to the values speed, shorter stride and step lengths, longer duration of
obtained for adults. However, linear displacements, veloci- double-support periods, and smaller ratios of swing to
ties, and accelerations were found to be consistently larger support phases,20,94,100 all of which are thought to increase
for these children than they were for adults.89 stability. Himann and associates94 found that between
Sutherland and colleagues,90 who studied 186 children 19 and 62 years of age, there was a 2.5% to 4.5% decline in
from 1 to 7 years of age, suggested that the following five the normal speed of walking per decade for men and women,
gait parameters could be used as indicators of gait maturity: respectively. After age 62, there was an accelerated decline in
duration of single-limb support, walking velocity, cadence, normal walking speed, that is, a 16% and 12% decline in
step length, and the ratio of pelvic span to ankle spread walking speed for men and women, respectively.94 Winter
(inter-ankle distance measured during periods of double and associates,17,45 comparing fit and healthy elderly sub-
support, and indicative of base of support). Increases in all jects with young adults, found that the natural cadence in
of these parameters except for cadence are indicative of in- the elderly subjects was no different from that in young
creasing gait maturity. In Sutherland and colleagues’ study, adults but that the stride length of the elderly subjects was
the duration of single-limb stance increased from about significantly shorter and both stance time and the double
32% of the gait cycle in 1-year-olds to 38% in 7-year-olds support times were longer in the elderly subjects than in
(normal mean adult value is 40%). Walking velocity also in- young adults. A significantly higher horizontal heel velocity
creased steadily, whereas cadence decreased with age.90 at initial contact was also reported, which would increase
Beck and colleagues91 found that time and distance meas- the potential for slip-induced falls, and which occurred even
ures and GRF measurements depended on speed of gait and though gait speed in the elderly was lower than that of the
age of the child. Increases in height and age were the major young subjects. Kerrigan and associates100 found that older
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 560

560 SECTION 5 ■ Integrated Function

persons, in comparison with younger persons, had reduced indicators of independence and general health,108 but also
plantarflexion ROM, peak hip extension ROM, and in- have predictive value for future health. Self-paced gait speed
creased anterior pelvic tilt, and suggested that subtle hip is the most common outcome measure for gait and reflects
flexion contractures and plantarflexor weakness might be the ability to transport the body from one place to another
causes of the joint changes in elderly people. in a timely manner. Perry and colleagues109 have suggested
Differences in kinetic measures of gait have also been that an average self-paced walking speed of 0.4 m/sec is the
reported. Winter and colleagues reported less vigorous minimum criterion for limited community ambulation, and
push-off by ankle plantarflexors in power analyses, as did 0.8 m/sec for unlimited community ambulation. Daily step
Kerrigan and associates.100 Mueller and coworkers found counts are also a new and popular method of assessing walk-
that plantarflexor peak torque and ankle dorsiflexion were ing function. It has been suggested that less than 5,000 steps
interrelated.101 These authors suggested that walking a day reflect a sedentary lifestyle and more than 12,000 a
speed and step length might be improved by increasing an- highly active lifestyle.110 Mean values for healthy older
kle plantarflexor peak torque and dorsiflexion ROM.101 adults are 5,000 to 6,000 per day.111,112 Walking function
Bohannon and colleagues102 found that hip flexor strength has predictive value. The ability to walk 400 meters and
was one of the variables that predicted gait speed, although the time taken to do so have been found to be important
these authors did not test plantarflexor strength. Strength- predictors for mortality, cardiovascular disease, and mobil-
ening and conditioning programs aimed at changing kinetics ity disability in a large sample of community-dwelling older
typically show similar results to those of Lord and associ- adults.113
ates,103 who conducted an exercise program for women Walking is known to be important for bone health.
60 years of age and older. After the program, the authors For example, poorer walking endurance, as measured by a
found significant increases in cadence and stride length, as 6-minute walk test,108 or lower ground reaction forces
well as reductions in stance time, swing time, and stance during walking in people with stroke, has been shown to
duration. However, detraining may be substantial. Con- correlate with lower paretic hip bone density, a condition
nelly and Vandervoort104 showed a strength decline of that contributes to hip fracture risk. Fast walking also
68.3% one year after a quadriceps training program ended appears to have positive benefits: The osteogenic index,114 a
for a group of elderly persons having a mean age of reflection of peak vertical GRF (or body weight), the num-
82.8 years. The speed of self-selected gait in this group ber of loading cycles, and number of walking sessions, has
also declined by 19.5%. been shown to be higher in fast walking than in walking at
That a less efficient gait is typical of aging is certain, self-selected speeds,115 and may promote greater bone
although the importance of possible causes awaits further health.
research. Winter17 reported both lower push-off work by Although the particulars vary, it is clear that there are
the ankle plantarflexors and higher levels of absorption by both kinematic and kinetic differences in the gait of elderly
the knee extensors (K3) at the same time in the gait cycle. people. These can stem from two sources: degeneration of
Although several causes for the lower push-off are possible, strength and balance control system or adaptation to make
the pistonlike action of a strong push-off may be perceived gait safer. It is likely that both are responsible.
as more destabilizing, and avoided by older subjects. The
reduced amount of work done by the plantarflexors to Gender
generate the work of walking may itself explain much of the
inefficiency of the gait of older people, as shown by Kuo.105 The research regarding gender differences in gait is fraught
Using a simple theoretical model, Kuo persuasively showed with the same difficulties as found in gait research with re-
that using larger amounts of ankle plantarflexor work (A2) is gard to age. Variations among methods, technologies, and
much more efficient than using the other two muscle subjects used in various studies make it difficult to come to
groups, the hip extensors (H1) and hip flexors (H3), to gen- many conclusions regarding the effects of gender. When
erate the positive work of walking.105 Less plantarflexor differences in height, weight, and leg length between the
work and more hip work is typical of the older walker, and genders are considered, gender differences are not very
hence would be expected to be much less efficient. Note, great. Oberg and associates116 found significant differences
however, that the increased absorption by the knee exten- between men and women for knee flexion/extension at slow,
sors also reported by Winter17 would also produce a less normal, and fast speeds at midstance and swing. They found
efficient gait. Another possible cause of inefficiency could a significant increase in joint angles as gait speed increased.
be reduction in the kinetic-potential exchange, described For example, the knee angle at midstance increased from
earlier in the Mechanical Energy of Walking section in this 15° to 24° in men and from 12° to 20° in women. However,
chapter, but there is no evidence for this. However, there is Oberg and coworkers looked at only the knee and hip. In
mounting evidence for increased co-contraction of antago- another study, the authors looked at velocity, step length,
nist muscles in the gait of older subjects,106 which would and step frequency.20 Gait speed was found to be slower in
reduce efficiency. In addition, age-related changes appear to women than in men (118 to 134 cm/sec for men and 110 to
affect lateral balance, and the resulting compensations may 129 cm/sec for women), and step length was shorter in
also explain some of the increased energetic cost of walking women than in men. Kerrigan and associates117 found that
in older adults.107 women had significantly greater hip flexion and less knee
Walking ability has important implications for health in extension during gait initiation, a greater internal knee ex-
the older adult population. Measures of walking are not only tension moment in late stance, and greater peak mechanical
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 561

CHAPTER 14 ■ Gait 561

joint power absorption at the knee at that time (K3). Kinetic permitting as normal mechanics as possible. For example,
data were normalized for both height and weight. These one may wish to limit ankle plantarflexion in a child with
authors also found that women had a longer stride length in cerebral palsy. However, in so doing, one would prefer to
proportion to their height and that they walked with a encourage the active energy-generating activity of the ankle
greater cadence than did their male counterparts.117 plantarflexors (A2) during push-off. Formerly orthoses did
not permit any ankle motion, but more recent designs in-
Assistive Devices clude hinged orthoses that prevent excessive plantarflexion
by use of a posterior stop but permit the ankle to move into
Walking without the use of assistive devices (crutches, dorsiflexion in late stance. Therefore, a child with cerebral
canes, and walkers) is preferred by most people. However, palsy is able to generate some power through the plan-
such devices often are necessary either after a lower-extremity tarflexion that follows. There is some evidence of decreases
fracture, when the healing bone is unable to bear full body in energy costs with use of an orthosis composed of elastic
weight, or as a more permanent adjunct for a balance, to straps that assist the hip, knee, and ankle movements in per-
compensate for muscle deficiencies or for joint pain. Recall ons with stroke.124 Attempts have also been made to make
from Chapter 10 that a very small force at the hand is use of mechanical characteristics of a leaf-spring design to
needed to produce a large balancing moment about the return energy to the foot during push-off. Although good in
standing leg because of the long perpendicular distance concept, designs to date have failed to show return of en-
from the point of application of the force on the hand to the ergy.94 Recently a carbon-fiber-spring ankle-foot orthosis
hip joint center. You may wish to review the sections in and a carbon-fiber knee-ankle-foot orthosis showed prom-
Chapter 10 on the use of a cane and the accompanying ising results when tested on children with ankle plantarflex-
figures at this time. Canes have typically been used on the ion weakness caused by myelomeningocele or arthrogryposis.
side contralateral to an affected lower extremity to reduce The orthosis supplemented ankle power and work and en-
forces acting at the affected hip, the reason being that a abled longer strides in almost all participants.125
lower abductor muscle force on the affected side would be
necessary to balance the weight of the upper body during
single-limb stance if an upwardly directed force was pro-
vided by the hand on a cane at some distance from the hip CASE APPLICATION
joint center. Krebs and coworkers118 tested the effect of
cane use on reduction of pressure through the use of an
Gait Status
instrumented femoral head prosthesis that quantified con-
tact pressures at the acetabular cartilage. The prosthetic
head contained 13 very sensitive pressure-sensing transduc-
at Discharge
case 14–8
Recall that Ms. Brown was prescribed an ankle-foot orthosis
ers. The magnitude of acetabular contact pressure was to assist with providing an adequate support moment during
reduced by 28% on one transducer and by 40% on another stance phase and to prevent foot drop and the resulting
transducer in cane-assisted gait in comparison with unaided energy-costly hip hiking. However, if the orthosis was rigid, it
gait. The reduction in pressure at the hip coincided with would make it impossible for her to achieve any A2-S push-
reductions in EMG amplitude in comparison with the same off during late stance. Two options were considered: (1) a
pace in unaided gait trials. The authors concluded that the hinged orthosis that permitted unlimited dorsiflexion but had
use of a cane on the contralateral side apparently allows the a stop beyond 10° of plantarflexion, thus allowing limited
person to increase the base of support and to decrease mus- push-off; and (2) a flexible ankle-foot orthosis that narrowed
cle and ground reaction forces acting at the affected hip. posterior to the ankle, thus permitting a limited range of both
The hip muscle abductor force was reduced, and gluteus dorsiflexion and plantarflexion. Because it was hoped that
maximus activity was reduced approximately 45%. Similar use of the device would be temporary, the latter was chosen.
conclusions were reported by Neumann in an article that Gait reeducation included progressive training without the
includes clear biomechanical explanations and figures.119 orthosis as strength was gained, and at discharge, Ms. Brown
Recall from our Patient Case that Ms. Brown used a four- used her orthosis only for outdoor use. As her stability im-
point cane on her unaffected side, because her arm was also proved, she was able to progress from a four-point cane to a
affected. For her, the most important objective was to gain straight cane. When she was discharged from outpatient
balance and stability, not to reduce joint forces. treatment, she was walking at 0.55 m/sec.

Orthoses
The function of orthoses in gait is to alter the mechanics of ABNORMAL GAIT
walking. They are used to support normal alignment, to
prevent unwanted motion, to help prevent deformity, to Both quantitative and qualitative evaluations of gait are use-
reduce unwanted forces or moments,120–122 and, more ful for assessors of human function. The most important
recently, to augment joint power.123 Although a full discus- quantitative variable is gait speed, which has been shown to
sion is beyond the scope of this book, the student can be related to all levels of disablement.126 An individual’s gait
deduce the effects of various types of devices with knowledge pattern may reflect not only physical or psychological status
of the mechanics of gait. In all cases, the wish is to reduce or but also any defects or injuries in the joints or muscles of
prevent unwanted movement or undesirable forces while the lower extremities. In assessing an abnormal gait, it is
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 562

562 SECTION 5 ■ Integrated Function

helpful to separate the cause or causes of deviations into Abnormalities in the central nervous system are respon-
structural impairment(s) and/or functional impairment(s). sible for conditions such as Parkinson’s disease, stroke, and
Functional impairments can be further categorized by ge- cerebral palsy. These all produce characteristic gaits that are
neal cause: whether directly due to abnormalities of the easily recognized by a trained observer. The parkinsonian
muscular or nervous system or their control, by pain, or as gait is characterized by an increased cadence, shortened
a result of compensations/adaptations to the abnormalities stride, lack of heel strike and toe-off, and diminished arm
or pain. Although an extensive review is beyond the scope of swinging. The muscle rigidity that characterizes this disease
this book, commonly encountered examples will provide a interferes with normal reciprocal patterns of movement.131
framework for examining abnormal gait. Cerebrovascular incidents, commonly known as strokes,
produce gait abnormalies that usually include reduction of
Structural Impairment strength and power of muscles of one lower limb that are
usually worse distally. These deficits are frequently com-
These are structural malformations that are congenital, pounded by some degree of spasticity and deficits in motor
caused by injury or by structural changes occurring second- control. Children with cerebral palsy sustain brain lesions
ary to injury. that are similar to lesions in adults who have had strokes;
A common structural abnormality is leg length discrep- however, the children sometimes have both limbs affected,
ancy. Kaufman and coworkers127 undertook a study to and the immature system produces shortened muscles that
determine the magnitude of limb length inequality that do not keep up with bone growth. Conversely, children with
would result in gait abnormalities. Many minor limb in- cerebral palsy often show compensations that people with
equalities are found in the general population but many of stroke do not use.
these do not necessitate any particular treatment or inter- When the plantarflexors, which are the major source of
vention because they do not have any significant effects on mechanical energy generation in gait, are unable to gener-
normal gait. The authors concluded that a limb length dis- ate sufficient power, muscles at other joints may compensate
crepancy of 2.0 cm resulted in an asymmetrical gait and had and provide even more energy than is typical of normal
the potential for causing changes in articular cartilage. Song gait.4 For example, Winter132 found that individuals with
and colleagues128 evaluated neurologically normal children below-the-knee amputations and below-the-knee prosthe-
who had limb discrepancies of 0.8% to 15.8% of the length ses used the gluteus maximus, semitendinosus, and knee
of the long extremity (0.6 to 11.1 cm). The compensatory extensor muscles as energy generators to compensate for
strategies observed were equinus position of the ankle and loss of the plantarflexors. Olney and associates49 found that
foot of the short limb (toe walking), vaulting over the long in children with cerebral palsy who had unilateral plan-
limb, increased flexion of the long limb, and circumduction tarflexor weakness, the involved plantarflexors produced
of the long limb. Children who used toe walking had a only 33% of the energy generation, in comparison with the
greater vertical translation of the body’s center of mass dur- 66% produced in normal gait.
ing gait than did normal controls. The quadriceps is normally active at initial contact
Structural problems may be implicated in running injuries. throughout early stance when the GRFV is tending to
Increases in the Q-angle, tibial torsion, and pronation of the extend the knee. It is common for people with patellar pain
foot may contribute causes to patellofemoral syndromes. In to inhibit quadriceps contraction and produce a gait that
running, stresses are greater than in walking, and so there is appears kinematically normal. The quadriceps function is
an accompanying increase in the likelihood of injury. In a sur- easily compensated for if a person has normal hip extensors
vey of the records of 1,650 running patients between the and plantarflexors. The gluteus maximus and soleus muscles
years 1978 and 1980, 1,819 injuries were identified.129 The pull the femur and tibia, respectively, posteriorly, which
knee was the site most commonly injured in running, and results in knee extension. Additional compensation, if the hip
patellofemoral pain was the most common complaint. extensor and ankle plantarflexor activity is inadequate, may
At the foot, pes cavus and pes planus cause alterations in be accomplished by forward trunk bending and a rapid plan-
weight and may cause abnormal stresses at the hip or knee. tarflexion after initial contact. The forward shifting of the
In pes cavus, the weight is borne primarily on the hindfoot weight moves the GRFV anterior to the knee (at initial con-
and metatarsal regions, and the midfoot provides only min- tact and during the loading response period). It also may
imal support.130 In running, the metatarsals bear a dispro- force the knee into hyperextension and eliminate the need
portionate share of the weight. In pes planus, the weight is for any quadriceps activity. This tendency is often seen in the
borne primarily by the midfoot rather than being distrib- early weeks after stroke, if not corrected with gait training.
uted among the hindfoot, lateral midfoot, metatarsals, and
toes, as it is in the normal walking foot. The propulsive
phase of gait is severely compromised. Concept Cornerstone 14-5
Functional Impairment Paresis of Dorsiflexors
This group includes all causes in which the timing and/or The normal functions of the dorsiflexors in gait are (1) to
amplitude of muscle activity is abnormal, whether due to maintain the ankle in neutral so that the heel strikes the
abnormalities of the muscular or nervous system or their floor at initial contact; (2) to control the external plan-
control, pain, or compensations/adaptations to the abnor- tarflexion moment at heel strike; (3) to dorsiflex the foot
malities or pain. in initial swing; and (4) to maintain the ankle in dorsiflexion
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 563

CHAPTER 14 ■ Gait 563

during midswing and late swing. If these functions are ab- trunk lean to the affected side, similar to the compensation
sent, the following would be expected to occur: (1) The for hip joint pain. Conservative means of reducing forces in-
entire foot or the toes would strike the floor at initial con- clude weight loss, use of a cane, external rotation of the foot
tact; and (2) the amount of flexion at the hip and knee while walking, and use of inner soles that change the GRFV.
would have to increase to clear the foot in swing phase. Pain also appears to be a factor leading to an increase in
An orthosis that supports the foot, called an ankle-foot oxygen consumption. As pain increases, oxygen consump-
orthosis (AFO), is frequently used to avoid these problems. An tion has been found to increase.137
orthosis that incorporates an electical stimulation at the
appropriate time, called functional electrical stimulation Adaptation/Compensation
(FES), or functional neuromuscular stimulation (FNS), has
This group includes all causes of variations that occur when
also been shown to be effective.133
a structural or functional impairment is present and other
structures alter their pattern to adapt to the abnormal con-
ditions, or attempt to maintain gait function by compensat-
In patients with bilateral lower extremity paresis or paralysis, ing for the reduced function. Sometimes both are present.
such as that caused by a spinal cord injury, walking usually in- The human body is remarkable in its ability to compensate
volves the use of long leg braces and crutches. In this form of for losses or disturbances in function. Most of the compen-
gait, the trunk and upper extremity muscles must perform all of sations that are made are performed unconsciously, and if
the work of walking, and the energy cost of walking is much the disturbance is slight, such as occurs in excessive prona-
greater than normal. Functional electrical stimulation is cur- tion of the foot, the individual may not be aware that the
rently being used to activate the paralyzed lower extremity mus- gait pattern is in any way unusual. However, most compen-
cles so that these muscles can generate energy for walking. sations will result in an increase in energy expenditure over
However, the energy cost of functional electrical stimulation– the optimal amount and may result in excessive stress on
induced walking is still well above that of normal gait.134 other structures of the body.
Asymmetries of the lower extremities that result in gait
adaptations may have structural or functional primary
Pain
causes, such as contractures of soft tissues around the joints,
This group includes all causes of variations that are attrib- bony ankylosis, and muscle weakness or spasticity.
utable primarily to pain. All overuse injuries and most joint
pathologies fall into this category. It is common for people
with patellar pain to inhibit quadriceps contraction and
Example 14-1
produce a gait that appears kinematically normal. The
quadriceps function is easily compensated for if a person has One might see excessive plantarflexion at the ankle during
normal hip extensors and plantarflexors, a situation that is stance phase in a limb that is normal. The primary cause
described above with reference to stroke. could be in the other limb: for example, an inability to
Many gait variations can be seen in osteoarthritis.135 For clear the toes in swing phase as a result of inadequate knee
example, hip joint pain causes an individual to reduce the flexion. The excessive plantarflexion by the stance limb
level of contraction of the hip abductor, the gluteus medius would be an adaptation.
muscle, during single limb stance. This results in a typical
pattern called Trendelenburg gait. Normally, the gluteus
medius stabilizes the hip and pelvis by controlling the drop
of the pelvis during single-limb support. If gluteus medius
Example 14-2
activity on the side of the stance leg is reduced, the pelvis,
accompanied by the trunk, will tend to fall excessively to- A somewhat different example of adaptation could result
ward the swing side, which would result in a loss of balance. from a knee flexion contracture. When the affected extrem-
To prevent the trunk and pelvis from falling toward the un- ity is weight-bearing, the normal extremity will have diffi-
supported side, the individual may laterally bend the trunk culty swinging through in a normal manner, as it appears to
over the stance leg at initial contact, during weight accept- be too long. A method of “equalizing” leg lengths is neces-
ance, and through single stance. This trunk motion enables sary for the swing leg to swing through without hitting the
the person to bring the center of mass closer to the hip joint, floor. Extra plantarflexion during push-off of the affected
thus reducing the need for such a large hip abductor con- side, described previously, would be one means. In this case,
traction and the pain it would have caused (see Chapter 10). there is no structural or functional abnormality in the adap-
Knee joint pain is the principal clinical problem in 10% tive movement. Alternatively, the person could increase the
of people over 55 years of age with osteoarthritis of the amount of flexion at the hip, knee, and ankle of the unaf-
knee, and one quarter of these are severely disabled, with fected side. Again, the limb showing the adaptation has no
extensive gait limitations.136 As knee varus is commonly structural or functional impairment. Other methods that
associated with osteoarthritis, the internal knee abductor produce relative shortening of the swinging leg are hip
moment (often reported in the literature as the external hiking, or circumduction of the leg. Each of these compen-
adductor moment, or simply “adductor moment”) is also sations makes it possible to walk, although they increase the
high, and is associated with abnormally large forces across the energy requirements above normal levels.
knee. Efforts to limit knee forces in the frontal plane include
2362_Ch14-524-568.qxd 2/2/11 4:44 PM Page 564

564 SECTION 5 ■ Integrated Function

Many compensations for inadequate power generation abnormal stresses, and decreasing energy expenditure.
have been identified.51 For example, persons with hemipare- Sometimes the corrective action may be as simple as using
sis resulting from stroke frequently show greater than nor- a lift in the shoe to equalize leg lengths or developing an
mal power generation from ankle plantarflexors at push-off exercise program to increase strength or flexibility at the
(A2-S) of the unaffected limb (interlimb compensation) or hip, knee, or ankle. In other instances, corrective action
in hip extensors in early stance (H1-S) on the affected side may require the use of assistive devices such as braces,
(intralimb compensation). canes, or crutches. However, an understanding of the com-
plexities of abnormal gait and the ability to detect abnormal
SUMMARY gait patterns and to determine the causes of these devia-
tions must be based on an understanding of normal struc-
The objectives of gait analysis are to identify deviations ture and function. The study of human gait, like the study
from the norm and their causes. Once the cause has been of human posture, illustrates the interdependence of struc-
determined, it is possible to take corrective action aimed ture and function and the large variety of postures and gaits
at improving performance, eliminating or diminishing available to the human species.

STUDY QUESTIONS
1. What percentage of the gait cycle is occupied by 10. What is largely responsible for the abductor moment
the stance phase in normal walking? How does an at the knee in stance phase?
increase in walking speed affect the percentage of 11. What are the roles of the hamstrings in normal gait?
time spent in stance? Do they contribute to support and/or to power?
2. What percentage of the gait cycle is spent in double 12. What are the major muscle groups that contribute
support? How is double support affected by increases to the positive work of walking, and when in the gait
and decreases in the walking speed? cycle do their contributions occur?
3. Describe the subdivisions of the stance and swing 13. Why does walking faster than normal and walking
phases of the walking gait cycle. slower than normal usually result in increased energy
4. During which period of the gait cycle does maximum costs?
knee flexion occur? 14. Why do long double-support times usually result in
5. What are the approximate values of maximum flexion increased energy costs?
and extension required for normal gait at the knee, 15. What is the role of the quadriceps muscle during
hip, and ankle? walking gait?
6. How does the total range of motion required for 16. What is the function of the plantarflexors during
normal gait at the knee, hip, and ankle compare walking gait?
with the range of motion required for running and 17. What are the functions of the dorsiflexors in normal
stair gaits? walking gait?
7. What is the difference between an internal moment 18. How is the swinging motion of the upper extremities
and an external moment? related to movements of the trunk, pelvis, and lower
8. What is the concept of the support moment, and extremities during walking gait?
what major muscle groups are responsible? 19. Compare muscle action in walking gait with muscle
9. What moments are acting at the ankle, knee, and hip action in running gait.
at initial contact? Answer the same question with re- 20. Explain what would happen in walking and running if
gard to different gait events: foot flat, midstance, a person’s plantarflexors were weak. What compensa-
heel-off, and toe-off. tions might you expect?

REFERENCES 4. Rose J, Gamble JG (eds): Human Walking (ed. 2). Philadelphia,

Williams & Wilkins, 1994.
1. Rowe PJ, Myles CM, Walker C, et al: Knee joint kinematics in gait 5. Winter DA: Biomechanics of normal and pathological gait: Implica-
and other functional activities measured using flexible electrogo- tions for understanding human locomotor control. J Motor Behav
niometry: How much knee motion is sufficient for normal daily life? 21:337, 1989.
Gait Posture 12:143, 2000. 6. Halliday SE, Winter DA, Frank JS, et al: The initiation of gait in
2. Sutherland DH: The evolution of clinical gait analysis. Part I: young, elderly, and Parkinson’s disease subjects. Gait Posture 8:8, 1998.
Kinesiological EMG. Gait Posture 14:61, 2001. 7. Sparrow WA, Tirosh O: Gait termination: a review of experimental
3. Sutherland DH: The evolution of clinical gait analysis. Part II: methods and the effects of ageing and gait pathologies. Gait Posture
Kinematics. Gait Posture 16:159, 2002. 22: 362, 2005.