This is an enhanced PDF from The Journal of Bone and Joint Surgery

The PDF of the article you requested follows this cover page.

THE MAJOR DETERMINANTS IN NORMAL AND

PATHOLOGICAL GAIT
J. B. dec. M. Saunders, Verne T. Inman and Howard D. Eberhart
J Bone Joint Surg Am. 1953;35:543-558.

This information is current as of May 13, 2008

Reprints and Permissions Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.
Publisher Information The Journal of Bone and Joint Surgery
20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org
‘.l’il 1 MA,JOI( i)ETEIIMINANTS iN NOItl\IAL ANI) PATIIOL( )(ICAL GAIT *f
BY J. B. I)EC. M. SAUNDERS, M.Ii., F.R.C.s.(EDIN.), VERNE T. INMAN, M.D., AND

HOWARD D. EBERHART, MS., SAN FRANCISCO, CALIFORNIA

From the Departments of A natomy and Orthopaedic Surgery, School of Medicine, and
The College of Engineering, University of California, San Francisco

INTRODUCTION

Human locomotion is a phenomenon of the most extraordinary complexity. Although

the primaty objective of locomotion may be simply stated as the translation of the body
from one point to another by means of a bipedal gait, its analysis requires the collection of
an enormous amount of data in order to follow the entire cycle of events. it is evident that
a complete (lescription of locomotion involves consideration of both the kinematics an(l
kinetics of the extremities in all their manifold details; but such knowledge, even if com-
plete, would be of little value to the orthopaedic surgeon unless it were integrated to evolve
a concept of locomotion from which deductions can be drawn and applied to the analysis
of the clinical problems which daily confront him. However, the synthesis of all the dc-
ments which simultaneously participate in locomotion, although an ideal worthy of
achievement, is a task of such magnitude and difficulty that its early attainment cannot be
expected. Therefore, it is our purpose to limit ourselves in this paper to a consideration of
what we have called the primary determinants of human locomotion and to examine these
determinants in relationship to pathological gait. It is our expectation that, by an appre-
ciation of these fundamental determinants, the orthopaedic surgeon will be able to analyze
disorders of locomotion with greater precision and to apply corrective measures with a
fuller understanding of the interrelationships which exist between the various segments of
the locomotor mechanism.

METHODS

The data upon which the present analysis of the fundamental determinants of gait is
based were obtained with the aid of a variety of techniques. Since a study of locomotion
involves the recording and the measurement of the magnitudes, directions, and rates of
change of the translations, rotations, and forces occurring in the body with respect to the
three coordinate axes in space, a number of methods must be employed in order to obtain
the different types of data. No one technique will yield all the desired information. The
choice of a particular method is dictated by such factors as the simplicity of recording, the
ease of reduction of the data, and the accuracy of the findings. The details of the tech-
niques employed have been described elsewhere 2 and, for reasons of space, cannot be
discussed here; but they will be briefly mentioned. The displacements of the extremities
were determined by several methods. One of these consisted of a glass walkway which per-
mitted the study of the displacements in the three planes of space simultaneously. By
measurement of the displacements as a function of time, the velocities and accelerations
of selected points were calculated. Displacements in the sagittal plane were also obtained
by an interrupted-light technique similar to that employed by Marcy. This method per-
mitted the calculation of the components of velocity and acceleration in the plane of
progression with great accuracy. Electrical accelerometers were used to check the findings
but were found to be inferior to methods of graphonumerical differentiation. The trans-
verse rotations of the limb segments were most accurately determined by the insertion of

* Based on two papers read at the Joint Meeting of the Orthopaedic Associations, London, July 1, 1952.

t A contribution from the Biomechanics Group aided by grants from The National Foundation for
Infantile Paralysis, Inc., and the National Research Council, Advisory Committee on Artificial Limbs.

VOL. 35-A, NO. 3. JULY 1953 543

544 J. B. D1:C. ‘SI. SAUNDERS, V. T. INMAN, AND H. D. EBERHART

pins into the bones. High-speed motion-picture photography proved to be of qualitative

value only, bUt it w-as useful for the examination of motions of minor magnitude. In some
instances Cinertt(hiOg1’aj)h had a limited usefulness. The gravitational and muscle forces
were studied l)V means of elect romyography and by the use of force plates. Electromyog-
raphy was most valuable in (letermining the precise phase of muscle activity. The force
plate, which will be mentioned in greater detail later, allowed the direct measuring from
the ground reactions the magnitu(les of vertical force, torque, and horizontal shears, as
well as the center of pressure on the foot. In addition special techniques were devised to
determine physical constants, such as the centers of gravity, the relative masses of the
segments, and the calculation of the mass moments of inertia. The data presented are for
level walking at the cadence normal to the subject investigated.

TilE CENTER OF GRAVITY OF THE BODY

In order to 1)rCSeIit a simplified concept of the phenomena of locomotion, it is con-

venient to consi(ler the l)ehavior of the center of gravity of the body during the cycle of
motion. By so (loing we introduce an abstract conception, since we assume that the entire
weight of the body is concentrated at that point. rfhe limbs may be regarded, therefore, as
weightless levei-s of the body. This enables us to avoid for the most part the discussion of
the forces acting upon the individual segments and to limit consideration to those oper-
ating at or influencing the center of gravity itself. In this way, we are able to treat of
locomotion in a somewhat idealized fashion which permits the analysis and synthesis of
the fundamental features concerned in the translation of the body and its energy require-
ments. Furthermore, the selection of the center of gravity as a point of reference from
which to depart possesses distinct advantages since the displacement pattern of the center of
gravity may be regarded as constituting the summation or end result of all forces and motions
acting upon and ConCcrfl((i with the translation of the body from one point to another during
locomotion.
Lnder static conditions, the center of gravity of the body has been determined by a
number of methods. Its position was found in cadavera by the familiar experimental tech-
nique of (louble suspension and in living subjects by the use of a balanced beam and by
calculation from volume contour maps of the body. The results from these several methods
are all in reasonable agreement. In adult males and females, it is estimated that the center
of gravity of the body lies in the mid-line at a distance from the ground corresponding to
about 55 per cent. of the total stature (individual variations being 1.25 per cent.) 1

With reference to the vertebral column, the center of gravity occupies a position just
anterior to the second sacral vertebra.

The Displacement of the (‘enter of Gravity

The rhythmic upward and downward displacement of the body is a feature of loco-
motion which is familiar to all. It may be seen subjectively in the upward and downward
fluctuation of distant objects on which the vision is fixed, and objectively in the “bobbing”
up and down of the head of a person, very noticeable when he is walking out of step in the
company of others. These fluctuations reflect the displacement of the center of gravity in
the vertical plane. The pathway described by the center of gravity in the plane of progres-
sion is a smooth undulating or sinusoidal curve. The form of this curve may often be seen
when a small boy holding a piece of chalk against his body at approximately the level of
his center of gravity scribbles a wavy line on a wall or fence as he walks parallel to it (Fig. 1).
The laboratory methods of recording the displacement of the center of gravity are no
more than an elaboration of this childhood pastime. The position of the center of gravity is
marked upon the body and the subject walks along a wall on which a coordinate system
has been drawn. He is photographed by synchronized cameras from different aspects and,
after correction for parallax, the displacements are determined in the various axes and

THE JOURNAL OF BONE AND JOINT SURGERY

 DETERMINANTS IN NORMAL AND PATHOLOGICAL GAIT

-- - r. ! . T 1’!

. I
‘ . . .., .
I -

: “‘1In

FI(. 1
Tine 1inc dna vn on i tin glass wino i( (W n’(pn’(’sen it t ia ni (1 )n’( )Xi ninat c )a I ii way [ t Iu. it cn’ ol
gravity of ti hO(lV.

a ic ielt I e( I to I Ii e l)iecise l)l iise ( )t t lie loc )Iiiot ( )I (yUle Ui( I I0 t lie )i I mat e ( )f t inic.
I ii flOilUal level “alking, t he (elit en’ of giavit y of I he body (lesciihes a siiioot Ii iegtilri.r
SiIHtSOi(lItl Ci1i\( iii t lie i)lt11e of progression. ii’oii I his (‘hiVe, it is (let ermilie(l thiit t lie
(cut (‘1’ of gravit y is (lisl)hitce(l t \\‘ice ill IL vent iCill (hile(t ion (luring I lie cycle of mot ion from
t he posit ion of t he lied-st rike of one foot t 0 t lie subsequent heel-st rike of t he same foot,
t lint is, as t he body 1)iL55C5 sti((essivel\’ in a double pitce over first the right limb and then
t he left . Fhie total aniount of this vent ical (hisplacemelit in noimtl a(lult males is about one
Ui(l eight -1 elit his inches ; ifl(hivi(hlal vatiat ions aie so snall that they nay be neglected.
‘Fhie sunnit S of these oscillrtt ions occili’ at 25 trl 75 i)er (cut . of the (‘(le, each corre-
s)onding t 0 the flhi(l(lle of the stance phase of the supporting linb ; the opposite linib is at
I his t iflie iii t lie Iiii(l(lle of t lie swing l)hIi5e. At 50 per cent . , or the middle of the cycle, the
(P1st er of gra\’it y falls to it s lo’est level ; I his position corresponds I o t he inteival of doul)le
‘eighit-beuing ‘hien bot Ii feet are in (unit rtct with the gi’oufl(l. As the cuive is followed in
iI S t15c(Iit ari(l (lescent , it is foiiil t o fltict tint e evenly bet veen t lie maxima and minima of
I he displacement, wit hi few if any irregularities. At its maximum vertical displacement, the
center of gravity of the body always lies slightly below the level of the same center when
tine subject is standing. in other words, a person is slight lv shorter when he is walking than
when lie is standing, so I hat, if lie were to walk through a tunnel the height of which corre-
51)011(15 eXactly to his standing height, hie could (to so wit hiotit fear of humping his head, as
Inc would hiave nearly one-half inch of clearance.
The center of gravity of t he ho(lv is also (lisplace(l laterally in the horizontal plane.
Rela live 1 o I he plane of progression, t lie center of gravit y (tescrih )es a sinusoidal curve, the
summits f which alternately j)it5S to the right and to the left in associat iofl Wit hi I lie sup-
pont of t lie ‘eight -hearing ext remit V. the curve is smoot lily undulating wit bout irregu-
larit les and it is similar in form to thiat of tine vertical displacement. The magnit tide of the
horizontal displacement in normal Irvel walking is approximately one and three-quarter
inches as measured from the extremes of the deviation from right to left.
When the vent ical and horizontal displacements of the center of gravity of the body
are combined and are projected on the coronal plane, thiey are found to describe an almost
perfect figure of eight, occupying approximately a two-inch square, since the vertical and
horizontal deviations are almost equal.

VOL. 35-A. NO 3, in’nx 1953

 .546 .1. ii. n)1C. ‘U. sAt’NI)ERS, V. T. INMAN, AND 11. D. EBEHIIART

.: thlcc-(IiflieflsiOflltl l)icture of the I)athwaY of the center of gravity in the line of

progression is that of a spiral alolig which the center moves with almost eveii velocity
( lig. 2 ). I in a single cycle ( )t loconot ion froni tine iight heel-strike t o tine left t oe-oft, I lie
((iit (i ol gia \‘ ily iiiL V I c It )ll O\V((l as ii rises sinoot lily to a suniiit 25 per ccii I of
at . the
cycle ; hu t ti t t lie sn ie I i inc it. (le’iItt es I 0 t lie right . As tine cent en i)egins to desceIl(l , it

smoothly deviates towar(1 the left to continue this deviation as it once more begins to
1.ise to a secon(1 summit at 73 per cent . of the cycle. When the second summit is reached
and tue center procee(ls to fall, it begins to return gradually to the right and has reached
tine central axis of tine forward motion as the cycle is completed.

THE DETERMINANTS OF GAIT

Newton’s first law of inotion states that every body continues in its state of rest or of
uniforni itiot iOI) iii It st n-night line, unless it h)e compelled h)y impressed force to change that
sI at e. ‘i’ranslat ion of a i)o(1\’ in a straight line with the least expenditure of energy maybe
achieved mechanically by

r - -- - ---- _ -- . means
quite
of the
impossible
wheel,
by
but
means
it is
of
bipedal gait. The next most
economical method would be
! translation of the body, as in
locomotion, through a sinus-
oi(ial pathway of low ampli-
ttide in whichi the deflections
are gra(iual. Since foice is
equal to mass times accelera-
tion (F = ma) an(l ttccelera-
tion is a. function of time
(a = v, t) , ai)i-upt changes in
the direction of motion com-
pd a high expen(hiture of
energy. In translating the
center of gravity through a
smooth undulating pathway
of low amplitude, the human
1)ody conserves energy; and,
as we shall see in considering
pathological gait, the body
l’o;. 2 will make every attempt to
TIn in(I(n’se(’ti((ni of (lie inon’izonital unIt thU vertical displarenients continue to conserve energy.
J)n’O(IU((S I lie pat iiwav of I he oniten’ of gravity no locomotion.
Developmentally, the achieve-
ment of this pat t ciii of progression would appeal- to have come about by natural experi-
ment . In childreii the pattern of locomotion is variable and irregular, and it does not be-
come stable until ap)i-0ximitte1y the seventh year of life. Therefore, it is of importance to
analyze the factors vhiichi determine the pathway of the center of gravity, since from them
we may gain considerable understanding of the locomotor mechanism and appreciate the
influence of pathological ahnoi-mahity, especially as regards the energy cost to the i)ody.
In order to (10 50, it is necessary to introduce at first certain purely theoretical considcra-
t i4)Ii5.

(viii pa.s. Gait

For analytical purposes, let us consider the behavior of the center of gravity in a
bipedal system in which the lowci- extremities are represented by rigid levers without foot.

THE JOURNAL OF BONE AND JOINT SURGERY

 DETERMINANTS IN NORMAL AND PATHOLOGICAL GAIT 547

Fin. 3
In the hypothetical compass gait, the
pathway of the center of gravity is a series
of ar(s.

Fin. 1

The i nfluenn-e of pelvic rot :ution flat (emns

t he arcs of t he pat hw:u of line centen’ of’
gravity.

VOL. 35-A, NO. 3, JUn.Y i953

548 j. B. DEC. M. SAUNDERS, v. T. INMAN, AND H. D. EBERHART

ankle, or knee mechanisms but are articulated at the equivalent of a hip joint. This permits
flexion and extension only (Fig. 3). In such a system quasilocomotion would produce
something analogous to the process of stepping-off distances with a pair of compasses or
dividers. The pathway of the center of gravity of the system in forward translation would
be a series of intersecting arcs. The radius of these arcs would be equal to the length of the
levers representing the extremities and, with each step, the angular rotation at the hip in
flexion would equal that in extension. Locomotion of this type might be imitated, but im-
perfectly, by walking on the heels with the knee fixed in extension.
The energy cost to a person using the compass gait would be exceedingly high. In
proportional terms of the average human stature and the length of stride, the center of
gravity of the body would have to be elevated approximately three and three-quarter
inches, which is double the normal vertical displacement. At the point of intersection of
the arcs, the ai)rupt change in direction of the forward acceleration would require the ap-
plication of force of considerable magnitude. Nonetheless it is from a consideration of such
a simple model that the determinants of locomotion can be followed.

The First Deteriii in a at: Pelvic Rotation

In normal level walking, the pelvis rotates alternately to the right and to the left,
relative to the line of progression. At the customary cadence and stride of average persons,
the magnitude of this rotation is approximately 4 degrees on either side of the central axis,
01 a total of some 8 degrees. Since the pelvis is a rigid structure, this rotation occurs alter-
nately at each hij) joint wiiich passes from relative internal to external rotation during the
stance phase.
The significance of pelvic rotation can best be appreciated by a study of our theoreti-
cal model (Fig. 4) . The effects of pelvic rotation are to flatten somewhat the arc of the
passage of the center of gravity in compass gait by elevating the extremities of that arc.
In consequence, the angles of inflection at the intersections of successive arcs are rendered
less abrupt and, at the same time, are elevated in relation to the summit. In this way, the
energy cost in locomotion is greatly reduced. The loss of potential energy is more gradual,
and the force required to (-hiange the direction of the center of gravity in the succeeding
arc of translation is less. The angular rotation at the hip in flexion and extension is re-
(tuced, an(l the energy required for the internal oscillation of the member is conserved.

‘J’lie Second Determinant: Pelvic Tilt

In normal locomotion the pelvis is tilted or lists downward relative to the horizontal
plane on the side opposite to that of the weight-bearing limb (positive Trendelenburg).
The alternate angular displacement is on the average 5 degrees. The displacement occurs
at. the hip joint producing an equivalent relative adduction of the extremity in the stance
phase and relative abduction of the extremity in the swing phase of the cycle. To permit
pelvic tilt, the knee joint of the non-weight-bearing limb must flex to allow clearance for
the swing-through of that member.
The effects of pelvic tilt on the pathway of the center of gravity are evident in the
experimental model
(Fig. 5). As the lateral list occurs while the body is passing over the
vertical supporting member, the center of gravity is lowered. Thus the summit of the at-c
of its translation falls and still further flattening of the pathway results. The saving of
energy by the cutting of the vertical displacement of the center of gravity in half is obvi-
ous. Furthet- energy is saved i)y greater reduction at the inflection of the arcs. The flexion
of the knee of the limb in the swing phase likewise conserves energy by the shortening of
the pendulum.

The Third Determinant: Race Flexion in the Stance Phase

A characteristic of locomotion is the passage of the body weight over the supporting

THE JOURNAL OF BONE AND JOINT SURGERY

 DI-:’rF:HMIN.NT5 IN NORMAI A ND PATIIOL0GIC.L (;AIT 549

1-in. .)

The effects of pelvic tilt on tin 10)11-

weight-hearinng side furl her flatten I he arc
of translation of t he ceniten’ of gravit v

Inn. (
Note I hat tine effects of knee flexion
(onni)ined ‘it Ii pelvic rot at ion and pelvic
tilt achieve minimal v(-rt i(Il (hisplacem(-n I
of the center of gravity.

VOn. 35-A, No. 3, JUlY 1953

550 J. Ii. DEC. M. SAUNDERS, V. T. INMAN, AND H. D. EBERHART

Fin. 7
The net effect of the determinants, pelvic
rotation, pelvic tilt, and knee flexion, produces
flattening of the arc of the center of gravity. This
is the equivalent of progressive lengthening of
I he extremity with tine same length of stride and
reduction inn the range of flexion and extension of
the iIiJ) as exempliho-oi inn the series of figures.

extremity while its knee joint is tin-

dergoing flexion. The supporting
member enters the stance phase at
heel strike with the knee joint in full
extension. Thereafter the knee joint
begins to flex and continues to (to so
until the foot is flat on the ground.
The average magnitude of this flex-
ion is 15 degrees. Immediately ante-
nor to the middle of the period of
full weight-bearing, the knee joint
once more passes into extension
which is immediately follow’ed by

the terminal flexion of tue knee.

This begins simultaneously with heel
--- --- rise, as the limb is at length carried
Fin, S
into the swing phase. This erioi of
The effects of the arcs of foot and knee rotation smooth the stance phase, occupying ai)out
out the abrupt inflexions at the intersection of the arcs 40 per cent. of the cycle, is referred
of translation of the center of gravity.
to as the period of “double knee
lock “, since the knee is first locked in extension, unlocked by flexion, and again locked in
extension prior to its final flexion.
If we examine the model (Fig. 6), the effects of walking over a flexed knee l)ecome
apparent. The summit of tue arc described by the passage of the center of gravity is still
further low-ct-ed with the conservation of energy by reduction of its vertical displacement.
These three determinants of gait-pelvic rotation, pelvic tilt, and knee flexion-all
act in the same direction by flattening the arc through which the center of gravity of the
i)O(ly is translated. The first-pelvic rotation-elevates the extremities of the arc, and the
second and third-pelvic tilt and knee flexion-depress its summit. The net effect is the
passage of tine center of gravity through a segment of a circle, the radius of which is two
and two-tenths longer than the length of the low-er extremity (Fig. 7). The relative length-
ening of the extremities reduces materially the range of flexion and extension at the hip

THE JOURNAL OF BONE AND JOINT SURGERY

 DETERMINANTS IN NORMAL AND PATHOLOGICAL GAIT 551

FIG. 9
The influence of the several determinants of gait on the passage of the center of gravity is seen in this
. ‘ exploded “ view of locomotion in which each stage of motion is equally spaced.

joint required to maintain the same length of stride. The relative elongation of the ex-
tremities plays an exceedingly important role in permitting increased velocities of gait at
slight increases in energy cost, since greater velocities of locomotion are achieved by the
lengthening of the stride rather than by increases in cadence.

The Fourth and Fifth Determinants: Foot and Knee Mechanisms

In the discussion of the fourth and fifth determinants of locomotion, we are concerned
with the smoothing out of the pathway of the center of gravity in the plane of progression
at the point of intersection of its arcs. Accurate recordings of the phase relationships of the
angular displacements at the ankle, foot, and knee show that their motions are intimately
related. Two intersecting arcs of rotation are established at the foot during the stance
phase (Fig. 8). The first arc occurs at heel contact and is described by rotation of the
ankle Itl)OUt a radius formed by the heel. The second arc is made by the rotation of the
foot about a center established at the fore part of the foot in association with heel rise.
At heel contact the foot is dorsiflexed and the knee joint is fully extended, so that the
extremity is at its maximum length and the centei- of gravity has reached its lowest point
of downward displacement. Rapid plantar flexion of the foot , associated with the initiation
of knee flexion, maintains the center of gravity with its forward progression at approxi-
mately tue same level for some time, thus flattening and slightly reversing the curvature
at the commencement of its arc of translation. The termination of this arc is similarly
flattened and slightly reversed by flexion of the second knee associated with heel rise. The
effects of the rotations at the foot on the displacement pattein of the knee and the (-enter
of gravity are illustrated in Figure 8. The obliteration of the abrupt inflexions at. the point
of intersection of the arcs of translation of the center of gravity smooths the gait by
establishing a sinusoidal pathway for its progression with a great reduction of energy cost.
The translation of the center of gravity in the plane of progression is illustrated in an
“exploded” view (Fig. 9).

The Sixth Determinant: Lateral Displacement of the Pelvis

As has been mentione(I, the center of gravity of the body is displaced laterally over
the weighit-hearing extremity twice during the cycle of motion. This displacement is pro-
duced by the horizontal shift the pelvis
of or by relative adduction at the hip. If the
extremities were parallel to oneanother, the amount of this displacement would neces-
sarily be half the interval between the axes of the hip joints, that is, approximately thiree
inches (Fig. 10). Excessive lateral displacement is corrected by the existence of the
tibiofemoral angle (Fig. 11) which, together with relative adduction at the hip, reduces the
(hiSplacement to about one an(1 three-quarter inches so that it approximates that of tine
vertical displacement. Thus the (leviation of the center of gravity is almost symmetrical
in both horizontal and vertical planes. Togethei, these deviations of the center of gravity

VOL. 35.A. NO. 3, JULY i953

552 J. fl. oN’. .\I. SAUNDERS, \. T. INMAN, AND H. D. EIIEHLIA RI’

Fit;. 10
Fig. l(): If I lie linnins were pan’:nllel, then’e woulol
in’ exo-o-s.sivo’ laten-al olisplao-ennennt of tine centen- Fin. 11
of gravity.
Fig. 11: Thurougin tine influence of’ a tii)iofennoral angle and of adduction at tine hip joint, exo-essive
I:nt en’al displ:n-ennenni is (on’recte( I

Fin. 12

The sum of tine (‘fleets of tine several determinants on the pathway of tine (‘enter of gravity
is viewed inn I nw’ phase relationship.

1’HE JOURNAL OF BONE AND JOINT Sl’RGlRY

 DETERMINANTS IN NORMAL AND PATHOLOGICAL GAIT .)33

contribute to the establishment in the plane of progression of the spiral of approximately

equal axes and add smoothness to the gait, since all sharp inflexions are obliterated and are
replaced by a sinusoidal curve. The sum of these factors is illustrated in true phase rela
tiOflshii1) in Figure 12.

THE CLINICAL EXAMINATION OF GAIT

Individual variations in locomotion are due to exaggerations in one or another of

these seven-al determinants. Owing to the interaction between the various factors, exag-
gerations in the range of one determinant are compensated for by reductions in another,
so that the final pathway of the center of gravity remains essentially the same in that it is
the most economical to maintain. Even gross pathological disturbances of the limbs have
little influence on the fundamental pattern, since the loss of one function is frequently well
compensated for h)V the exaggeration of another. Therefore in the clinical examination of
(hisOr(ters of gait, the attention should be focused serially on the six determinants already
described, an(l the exaggerations and losses in their range should I)e noted. In this way
even minor disturi)ances of gait may be readily ltssesSe(l.

Evaluation of Patholoqical Gait: Energy Levels

Before pathological gait is considered, something should be said coneeining the energy
levels (luring locomotion. In or(ler to aVOi(l lengthy mathematical formulation, qualitative
teims will 1)e laigely employed. rflie energy level of a body is tile sum of its potential and
kinetic enet-gies. If no work is done, the energy level is constant, as in the classical cx-
ample of a simple pendulum where the loss of potential energy is exactly compensated for
by the gain in kinetic energy. When the energy level is not. constant, then work must ie
(lone in ordei to produce the change in energy level. In the computing of the energy levels
of the diffeient segments of the low-er extremity, it is found that the levels are not con-
stant. The difference is a measure of the work done by the muscles at the joints. The net
result is the forw-ard displacement of the body, but a large portion of the energy is dissi-
pated in the i-otations of the segments. These effects are referred to as the output an(1 in-
put of energy respectively.
From stu(hies of the energy levels, it has been establishied that the Outj)ut of the ankle
and hip is considerably greater than the input ; therefore, most of the energy required for
level walking is provided by the muscles acting on these joints. In the case of the knee
joint, the out}Jtit is so much less than the input that this joint predominantly absorbs en-
ergy. However, the knee (leereases the vertical motion of the body by flexion; and, al-
though energy is dissipated in the process, the overall energy requirement is less than
w-ould he necessary in walking over a rigid knee. Additional energy is ahsOrbe(l by tile knee
to decelerate the leg and foot (luring the swing phase. Nonetheless, not all of the energy
absorbed by the knee is lost. A considerable portion is stored and is returned to the system
in the later Part of the sw-ing phase by imparting continued forward acceleration to the
body at the time -hen most of its potential energy is lost. Thus locomotion is not only
(Inc to the “push” of the member in support hut also to the “pull” of the (leceleration at
the swinging knee.
The energy expended (luring straight and level walking at a constant cadence is
divided approximately equally between the production of rhythmic oscillations of the legs
and the elevation and depression of the center of gravity of the body. If the human
mechanism were a truly efficient machine, the kinetic energy in the system \o1il(l he con-
‘eited into Potential energy and would be stored until it is required to initiate movement
in one or more of the segments of the body. However, since tile muscles, when stimulated,
expend energy 1)0th dut-ing conti-act ion and elongation, energy storage is never complete or
recoverable. There is evidence, however, that under certain specific and rigid situations,
energy storage does occur in the muscular system and perhaps as much as 40 per cent. of

VOL. 35-A. NO. 3, JULY 1953

554 J. I). DEC. M. SAUNDERS, V. T. INMAN, AND H. D. EBERHART

200 - VERTICAL LOAD

/00 -

C FUSED ANKLE ao NORMAL ANKLE

2 TIME, SECONDS

Q_40 FORE AND AFT SHEAR

2\j/)/:t,v//’\ FIG. 13
I

Fig. 13: The vertical acceleration of the body

of a person with panarthrodesis of one foot is
shown. There is no appreciable difference be-
tween records of the fused ankle and the normal
ankle (upper figure). The lower figure illustrates
that the fore and aft shears are not altered.

Fig. 14: The abrupt inflexion in the arcs of

motion at the fused ankle is transmitted to the
knee joint which by excessive flexion maintains
the smoothness of the path of translation of the
center of gravity.

the energy put into the locomotor mech-

anism is recoverable. The factors which
permit this energy storage and recovery
will not be discussed here, but they in-
volve the precise timing of the muscle
contraction to the displacements of the
moving segments, the limiting of the
muscle contraction to very brief periods
Fin, 14
of activity, and the action of two joint
muscles.
As the center of gravity of the body is displaced through its sinusoidal pathway of low
amplitude, energy is expended during its elevation, and only a portion of this energy is
recovered on its descent. The net result is a continual expenditure of energy. The magni-
tinde of this- expenditure is qualitatively appreciated i)y the common experience that
walking up a slight grade is much more fatiguing than w-alking on the level, whereas walk-
ing down a gentle slope is still easier. A descending grade of approximately 4 per cent.
requires little or no elevation of the body against the pull of gravity.
With the loss of one of the major determinants as a result of injury or disease, the
i-emaining mechanisms will compensate w-ithin their ai)ility to produce an over-all displace-
ment of the center of gravity of the body through a sinusoidal path of low amplitude, since
this requires the minimal expenditure of energy. That this is true is confirmed 1)y Hfl

analysis of pathological gait.

THE JOURNAL OF BONE AND JOINT SURGERY

 DETERMINANTS IN NORMAL AND PATHOLOQICAL GAIT

The Force Plate VERTICAL FLOOR REACTIONS

A rea(Iy means is availai)le for tine

determining of tue vertical displacement of O.D. B.w.
tine center of gravity of the body thi-ough the
use of a force plate. A force plate is a device
for measuring the ground i-eactions, including
torque, (luring locomotion. Essentially the
al)Paratus consists of a heavy metal plate
which is rigidly supported by four metal
columns to which are bonded a series of Bare foot Flat shoes
opposing pairs of strain gages (some fifty in
all). The deformations in the x, y, and z
coordinates, as well as torque, are measured
by balanced 1)ridge circuits and are iecoided
OIl an oscillograph. In operation, two such)
force plates are employed; the subject Step-
ping, in the course of walking, first on one,
then on the other plate. ‘ihese are set flush) Flat hQels 3 heels
with the walkway. In essence, the apparatus
acts as an inertialess scale of great sensitivity.
The force plate records directly the vertical
accelerations. From these measured accelera-
tions, it is easy to calculate the displacement
of the center of gravity of the body by geomet-
nc integration. Furthermore, since the force #{149} ,, h I
plate measures accelerations, it is exceedingly eg s
sensitive to any changes in the normal dis- Effect of varyin9 heal hes9ht’5
placement pattern of the body as a whole on VertcaI Floor Reactions
and is more accurate than photographic
methods for the determination of sudden . . 1G. 15
. . . Varying heel heights have little effect on verti-.
deviations of the center of gravity of the cal floor reactions, indicating a smooth transla-
body caused by aitei-ations in acceleration. tion of the center of gravity.
The force plate, therefore, becomes a very
useful instrument for the study of pathological locomotion and for the evaluation of
the degree of compensation which has occurred following the loss of one or more of the
major constituents of the lower extremity. Minor changes in the displacement pattern
are amplified when they are recorded as changes in vertical acceleration.

PATHOLOGICAL GAIT

When a pci-son with a panarthrodesis of the ankle and subtalar joints walks, minor
changes in the displacement pattern of the knee and hip may be noted. These small
angular changes in the knee and hip compensate for the functions lost in the foot and
ankle. The degree of compensation may be measured by recording the vertical accelera-
tions of the force plate. In the force-plate record presented, the compensation is excellent;
the only abnormality is the shock wave noted during the transition of body weight from
the sound to the arthrodesed foot (Fig. 13). The pathway of the center of gravity of
the body is normal and is accomplished by exaggerating the initial knee bend on the side
of the arthrodesed joints. Because of the fixed ankle, the displacement of the knee joint
is over two intersecting arcs, the centers of which are located at the heel and ball of the
foot. However, an exaggeration of the initial knee flexion prevents the discontinuity of
the curve of knee displacement from being transmitted to the trunk and maintains the

VOL. 35-A. NO. 3, JULY 1953

 .1. H. n)n.;(’. \I. SAUNDERS, V. T. INMAN, AND II. D. EIIERHART

ZI-.
Oz
OW

o
0
-Jil.
11.0

Id Id
>Q.

120
PERCENT OF CYCLE

VERTICAL FLOOR REACTION FOR

SUBJECT WITH FUSED HIP

Fnn. 1

Tine (Iiffen’ennc(’ inn I lie floor neactionis in the presence of a fused hip is showni. These
al)nnorlnalities are refl(’(-te(l oIl the normal side anud are indicative of tine attempt to cOni)-
l)(n1sate at the nornnml inij) for tine loss of motioin on tine affected side.

I1OIlilal (lisplacenlent I)1t t(’IIl of tile (‘enter of gravity of the body (Fig. 14) . Thus the
(‘OnlIJIete less of the foot an(1 ankle mechanism has b)een completely compeflsate(l for
1)0,’ slight Illodificat iOilS iii the beliavion of tue knee and hip ‘ith retention of the normal
l)itt tern of (lisplacenlent ()I the ceiltel of gravity of tile body.
A siniilai- COflll)eilsat()Iv mechanism is vell (lepicted in the change in gait seen in
\VoIiTlen iiile they tie walking in shloes with various heights of heels. In low heels, the
angular displacement of t he vaiious segments of the lower extremity are similar in both
sexes. However, high heels (lecrease the contributions of the foot to the integrated whole
and this loss is made up by modifications in the remaining constituents, namely, the
pelvis, hips, and knees. Tine initial knee flexion is increased and the pelvic tilt and rotation
RIP (‘xaggeitite(i, pi-oducing the (-haractei-istic gait which is familiar to all. This is, how-
ever, not met-ely an aflecte(l gait, since the force-plate records show that in this gait
similar vei-t ieal aceelerat ions are achieved, whether the person is in high or low heels
(Fig. 15). The exaggel-at((l movements of the pelvis and the mci-eased hip an(l knee
flexion an-c eompensatoi-v motions to offset the loss’ of the foot mechanism cause(1 by
the weal-ing of high heels, with t.he result tilat a normal pathway of displacement of the
(‘emit ci- of gI-avity of the body is maintained.
\Vhen tine knee is immobilized, its important contm-ibution to the orderly and harmoni-
Otis ai-i-angement of the segments of tile lower extremity obviously is lost. The limp which
chai-aeten-izes a patient with a stiff knee requires no detailed description, but certain
of tine compensatoi-y movements in the remaining normal segments should be pointed
out. To gain clearance foi- the foot, excessive elevation of the pelvis on the affected side

THE JOURNAL OF BONE AND JOINT SURGERY

 DETERMINANTS IN NORMAL AND PATHOLOGICAL CAIT 557

is necessary, or an increase in the heel rise is required on tile normal side. However, in
Sl)it(’ of these (leviations from what are considered as normal bodily motions, tile centci-
of gravity of the body retains a remarkably constant and nearly normal path of vertical
(hisplacelnent as noted from the force-plate records. rfili S particularly true if flexion at
t Inc kiiee of from is to 20 degrees is pieserved at the time of its fixation.
The nlechianisms of tile low-er extremity apparently are able to compensate reasonably
vell for tile stiff knee as far as the maintenance of the noimal pathway of the center of
gravity of the 1)ody and the minimizing of the energy loss necessary for external work.
Iloweven-, the energy necessary to initiate the swing phase on the side of the stiff knee
is increased to almost three times that of the normal. The flexion of the knee and, to a lessei
extent, tile flexion of the hip, which occurs late in the double weight-bearing phase and
(‘ally in the swing phase of walking, is produced partly by the upward thrust imparted
to tile leg 1)y plantar flexion of the foot and partly by the flexor muscles of the hip, the
latter contrib)uting approximately one third and the former, approximately two thirds.
Early flexion of the knee at the initiation of the swing phase is important, since it converts
t lie leg into a double pendulum and thus decreases the moment of inertia of the whole
lower extremity. This in turn decreases the force requirement of the flexors of the hip.
Fixation of tile knee suppresses this mechanism and requires the hip flexors alone to initiate
t lie swing pilase. This demands that a force be exerted by the hip flexors that is several
times greater than tile normal as may he observed by the hesitancy at the commencement
of the swing phlase in the gait of a person with an arthrodesed knee.
The impon-tance of the knee joint in minimizing the energy requirements in locomo-
ti()Il should be emphasized. The knee contributes little toward the propulsion of the
body thi-ough space, but the knee joint does function to reduce the energy necessary to
initiate tile swing phase and to smooth out the path of the center of gravity. Immobiliza-
tion of the knee by arthrodesis, casts, or long braces leads to almost tripling the energy
tequired to start the swing phase of the involved limb. A vivid example of the role played
by the knee joint in human locomotion is provided by the experience gained in the study
of prostheses for above-the-knee amputees. Much effort has been expended by the Ad-
visoi-y Committee on Artificial Limbs of the National Research Council to develop
fllecilanical and hydraulic devices to be incorporated into the prosthetic knee joints.
All of these mechanisms were intended to limit the heel rise at the beginning of the swing
1)llase an(1 to improve the appearance of the amputee’s gait.. However, the increased
iPsistance to flexion and extension in the knee joints, which the devices provide, requires
the amputee to expend more energy. (The devices were, therefore, disliked by the ampit-
tee.) Many an amputee would leave the limb shop with a well functioning device and an
improved appearance in his gait, only to return at a later time with a free-swinging knee
joint with its resultant abnormal heel rise. With a wrench and a screw driver the amputee
had managed to render the mechanism functionless so that the energy cost was lowered at
the expense of the aesthetic appearance of his gait.
The exponents of pylons for above-the-knee amputees realize the importance of
tine knee joint in decreasing tile amount of energy required to initiate the swing phase on
the amputated side w-hen they insist that a pylon should he as light as possible in weight.
The reduction of the -eight placed distally in the extremity does not appreciably increase
the moment of inertia over that of the stump itself and thus requires minimal effort to
stai-t tine swing phase. In tile conventional prosthesis, the conversion of the lower extremity
into a (louble pendulum by the incorporation of a free knee joint obviates, to a certain
degree, this difficulty.
Tile relatively inconspicuous limp caused by an arthrodesed hip in good functional
posit ion can easily be detected by abnormalities in the vertical floor reactions (Fig. 16).
[he remaining normal segments of the inferior extremity are able to compensate only
partially for the loss of the hip joint. It will be recalled that pelvic rotation and pelvic tilt

VOL. 35-A. NO. 3. JULY 1953

558 .. B. m)LC. M. SAUNDERS, V. T. INMAN, AND U. D. EBERtIAIt

are important factors in (lecreasing the vertical displacement of the center of gravity
of tile i)O(ly. At the point of maximal elevation of the center of gravity of the body, the
hip joints are in mid-position with respect to rotation. From this point to the position of
double weight-bearing, in which the feet are videst apart in the stride, the pelvis rotates
externally about the weight-bearing hip joint. This may be expressed as internal rotation
of the femur with respect to the pelvis. This movement, although of small magnitude, is
of considerable importance. When the hip is affected by disease with resultant limitation
of motion, it will be found that the hip vill be held in external rotation during the swing
phase so as to be in a position to make available during the stance phase what little internal
rotation remains and to allow the pelvis to rotate in an approximately normal manner.
The loss of one of the major joints of the extremity is almost fully compensated for
by exaggerated motions at other levels as studies in below-the-knee amputees clearly
reveal. In such cases the vertical floor reactions indicate that the center of gravity of the
body is still translate(l smoothly through a sinusoidal pathway at low energy cost. How-
ever, the amputee who has lost both foot and knee is unable to compensate satisfactorily
and the vertical floor reactions deviate markedly from the normal pattern. Because of the
minor changes in direction from the smooth sinusoidal pathway of the center of gravity,
high accelerations are required which dissipate energy and make such deviations very
costly. So great is the cost that, as our experience has shown, the loss of two joints in
the elderly subject will inevitably shorten life from the demands upon his cardiovascular
system which must supply his requirements at the usurious rate of 300 per cent.

SUMMARY

Human locomotion is a phenomenon of the most extraordinary complexity in which

so great arethe multitude of individual motions occurring simultaneously in the three
planes of space that analysis is difficult without some unifying principle. The adoption
of the concept that fundamentally locomotion is the translation of the center of gravity
through space along a pathway requiring the least expenditure of energy supplies the
necessary unifying principle which permits of qualitative analysis in terms of the essential
determinants of gait. The six major determinants are pelvic rotation, pelvic tilt, knee
and hip flexion, knee and ankle interaction, and lateral pelvic displacement. The serial
observations of irregularities in these determinants provides insight into individual varia-
tion and a dynamic assessment of pathological gait. Pathological gait may be viewed as
an attempt to preserve as low a level of energy consumption as possible by exaggerations
of the motions at unaffected levels. Compensation is reasonably effective with the loss
of one determinant of which that at the knee is the most costly. Loss of two determinants
makes effective compensation impossible and the cost of locomotion in terms of energy
is increased threefold with an inevitable drain upon the body economy.

REFERENCES

1. CROSKEY, M. I.; DAWSON, P. M.; LUESSEN, A. C.; HAHRON, I. E.; and WRIGHT, H. E.: The Height
of the Center of Gravity in Man. Am. J. Physiol., 61: 171-185, 1922.
2. EBERHART, H. D.; INMAN, V. T.; SAUNDERS, J. B. DEC. M.; LEvENS, A. S.; BRESLER, B.; and
MCCOWAN, T. D.; Fundamental Studies on Human Locomotion and Other Information Relating to
Design of Artificial Limbs. A Report to the National Research Council, Committee on Artificial Limbs.
Berkeley, University of California, 1947.
3. MARRY, E. J.: La m#{233}thode graphique dans les sciences exp#{233}rimentales, Paris, G. Masson, 1885.

THE JOURNAL OF BONE AND JOINT SURGERY