uvv ha?
..n....a.u.y.m
52f.

 

an.

.23

x .
w

fifiWfi-fi?
. a . . :oi
nuauwmw
£4".

1

 

h h. E
. .1. c.
ti. . l2.
1;: 3.x 3. .
5:39.30... v...
W. 1“!!!

. w ,

.1? .I

 

. v I .. , ill 4
III-Um 1.. :s‘...u.‘.e... . $5.31} ‘ .cQg‘H... é.9.3.&.§

 

   

. . 1.... 2. I

F:

  

1
5 in it ‘ LIBRARY
gamer" Michigan State

University

This is to certify that the
thesis entitled

Gait analysis of forelimbs in Thoroughbreds with
metacarpophalangeal joint injuries

presented by

Patricia Eliza de Almeida

has been accepted towards fulfillment
of the requirements for the

 

 

 

 

 

Master of degree in Large Animal Clinical Sciences
Science
//Z( 4 MG o
' Major Profesfir’ 3 Signature
m/a /o :3
Date

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIFiCflDateDuepSS-DJS

 

GAlT ANALYSIS OF FORELIMBS IN THOROUGHBREDS WITH
METACARPOPHALANGEAL JOINT INJURIES
By

Patricia Eliza de Almeida

A THESIS
Submitted to
Michigan State University
in partial fulfillment to the requirements
for the degree of
MASTER OF SCIENCE

Department of Large Animal Clinical Sciences

2003

ABSTRACT
GAIT ANALYSIS OF FORELIMBS IN THOROUGHBREDS WITH
METACARPOPHALANGEAL JOINT INJURIES
BY
Patricia Eliza de Almeida
This study aimed to determine biomechanical variables for the diagnosis of
lameness, and to use graphical representations of data to enhance the
understanding of the movement pattern in horses with fetlock joint lameness.
Five lame (group 1) and five sound (group 2) horses had kinematic and ground
reaction forces (GRF) data recorded at trot. Group analysis demonstrated that
the lame limb had significantly smaller (p<0.05) fetlock joint maximum extension
angle and vertical GRF peak (220°; 7.9N/kg) than the sound limb (239°;
10.8N/kg). Lame horses also showed longer (p<0.05) stance duration (49%) than
sound horses (43%). Due to high variability between horses, single-subject
analysis was used to detect intra-individual differences between forelimbs of
group 1 (eg. horse 2 peak braking GRF: lame = -1.4N/kg, contralateral = -
0.6N/kg; p<0.05). Graphical displays including kinegrams and force vector
diagrams facilitated identification of the lame limb versus the contralateral limb.
Variability of some kinematic and GRF measures was reduced in the lame
horses, which may be caused by compensatory mechanisms to reduce pain in
the affected limb. Biomechanical measures, including movement variability, may
assist clinicians in objectively assessing lameness, improving the understanding

of gait abnormalities and evaluating improvement in response to therapy.

Copyright by
PATRICIA ELIZA DE ALMEIDA
2003

DEDICATION

In loving memory of my grandmother Eulalia.

ACKNOWLEDGMENTS

My family for unconditional support and love.

Dr. Hilary Clayton for being my mentor and for giving me the opportunity to
work at the Mary Anne McPhail Equine Performance Center.

Committee members Dr. James Lloyd and Dr. John Stick for their input in
crucial stages of the development of this thesis.

Committee member Dr. David Mullineaux for help and patience with the
writing and edition of this thesis, for being a terrific researcher, tutor and
fnend.

Joel Lanovaz for technical support on data collection and analysis during
the preliminary phase of this study.

Dr. Adroaldo Zanella and Animal Behavior and Welfare Group for kindly
supporting me during the development of this study.

Jo Anne Normile and horses from CANTER for their generosity, kindness
and patience.

Kirsten and Connie Pepper for love, friendship and comfort during the

development of this thesis.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................. viii
LIST OF FIGURES ............................................................................................ ix
CHAPTER 1 ..................................................................................................... 1
INTRODUCTION ............................................................................................... 1
CHAPTER 2 ..................................................................................................... 5
LITERATURE REVIEW ..................................................................................... 5
2.1 HISTORY AND TECHNIQUES OF EQUINE GAIT ANALYSIS ......... 5

2.2 APPLICATIONS OF EQUINE GAIT ANALYSIS ................................ 10

2.3 MUSCULOSKELETAL INJURIES IN RACING THOROUGHBREDS17

2.3.1 Epidemiological aspects ...................................................... 17

2.3.2 Mechanisms of musculoskeletal injury ................................ 19

2.4 CLINICAL EVALUATION OF LAMENESS ........................................ 22

2.5 USE OF BIOMECHANICS IN LAMENESS EVALUATIONS ............. 25

2.6 GRAPHICAL PRESENTATIONS IN GAIT ANALYSIS ...................... 28

2.7 VARIABILITY OF MOVEMENT ......................................................... 32

RESEARCH OBJECTIVES ............................................................................... 36

HYPOTHESES .................................................................................................. 36

CHAPTER 3 ...................................................................................................... 37

METHODS OF INVESTIGATION ...................................................................... 37

3.1 Subjects ....................................................................................................... 37

3.2 Data collection protocol ................................................................................ 39

3.3 Data analysis ............................................................................................... 42

3.3.1 Statistical analysis ..................................................................................... 45

CHAPTER 4 ..................................................................................................... 47

RESULTS ......................................................................................................... 47

4.1 KINEMATICS ......................................................................................... 47

4.2 GROUND REACTION FORCES AND IMPULSES ................................ 55

4.3 BIOMECHANICAL VARIABILITY .......................................................... 60

4.4 GRAPHICAL REPRESENTATIONS OF BIOMECHANICAL DATA ....... 62

4.4.1 Kinematics .......................................................................................... 62

vi

4.4.2 Ground reaction forces ....................................................................... 64

CHAPTER 5 ..................................................................................................... 67
DISCUSSION .................................................................................................... 67
CHAPTER 6 ..................................................................................................... 77
CONCLUSIONS ............................................................................................... 77
REFERENCES ................................................................................................. 79

vii

LIST OF TABLES

Table 2.4.1 - Lameness grading score system .................................................. 24

Table 3.1 - Descriptive characteristics and lameness diagnosis of subjects from
group 1 (R = right; L = left) ................................................................................. 38

Table 3.2 - Descriptive characteristics of subjects from group 2 ........................ 38

Table 4.1.1 - Mean (SD) joint angles for lame (L) and contralateral (C) forelimbs
of the lame horses (group 1), and for the “mean sound" forelimb of the control
horses (S; group 2) ............................................................................................ 48

Table 4.1.3 - Mean (SD) stride duration, stride length and stance duration for
lame (L) and contralateral (C) forelimbs of lame horses (group 1), and for the
“mean sound” forelimb of the control horses (S; group 2) .................................. 54

Table 4.2.1 - Mean (SD) peak vertical, braking and propulsive GRFS for lame (L),
contralateral (C) forelimbs of lame horses (group 1), and for the “mean sound”
forelimb of the control horses (8; group 2) ......................................................... 55

Table 4.2.2 - Mean (SD) vertical, braking and propulsive impulse of lame (L) and
contralateral (C) forelimbs of the lame horses (group 1), and for “mean sound”
forelimb of the control horses (S; group 2) ......................................................... 59

viii

LIST OF FIGURES

Figure 2.1.1 - Chronographic method of motion analysis in a galloping horse
entiled “The horse in motion” ............................................................................ 6

Figure 2.1.2 - Gait analysis being performed using Motion Analysis System....8

Figure 2.2.1 Fetlock joint angle pattern of lame (left) and contralateral (right)
forelimbs using a model in which lameness was induced by pressure on the hoof
sole .................................................................................................................... 13

Figure 2.2.2 - Shoulder joint angle pattern of lame (left) and contralateral (right)
forelimbs ............................................................................................................ 14

Figure 2.2.3 - Mean (n = 6) vertical ground reaction force during the stance
phase for sound horses (mean values for 2 forelimbs) and lame horses (lame
and compensating forelimbs) with superficial digital flexor tendonitis ................. 15

Figure 2.2.4 — Mean (n = 6) longitudinal ground reaction forces during the stance
phase for sound horses (mean value for 2 forelimbs) and lame horses (lame and
compensating forelimbs) with superficial digital fiexor tendinitis ........................ 16

Figure 2.3.2.1 - Apical (left) and basilar (right) proximal sesamoid fracture ....... 20

Figure 2.3.2.2 - Suspensory breakdown injury with comminuted fracture and
distraction of both sesamoid bones .................................................................... 21

Figure 2.6.1 - Stick diagram of the forelimb of an individual horse trotting at a
velocity of 4m/s on a treadmill from left to right during a complete stride ........... 29

Figure 2.6.2 - Vector diagram (or vectordynamogram) of mean (n = 4) Fy-Fz
force-curves of all limbs during control session (HO) and moderate left forelimb
lameness (F3) .................................................................................................... 31

ix

Figure 3.2.1 - Coordinate system used during GRF and kinematic data collection
............................................................................................................................ 40

Figure 3.3.1 - Latero-medial radiograph of the hoof showing schematic
determination of the coffin joint center of rotation .............................................. 43

Figure 4.1.1 - Fetlock joint angle for the “mean lame” (solid gray line) and “mean
contralateral” (gray line with circles) forelimbs of the lame horses (group 1), and
for the “mean sound” (solid black line) forelimb of control horses (group 2) ...... 49

Figure 4.1.2 - Fetlock joint angles for lame (solid gray line) and contralateral
(gray line with circles) forelimbs of the lame horses (group 1) ........................... 50

Figure 4.1.3 - Carpal joint traces for “mean sound” (solid black line) and the 5
trials for lame (solid gray lines; above) and contralateral (gray lines with circles;
below) forelimbs of lame horses (group 1) ......................................................... 51

Figure 4.1.4 - Coffin joint traces for “mean sound” (solid black line) forelimbs of
the control horses (group 2) and the 5 trials for lame (solid gray lines; above) and
contralateral (gray lines with circles; below) forelimbs of the lame horses (group
1) ........................................................................................................................ 52

Figure 4.1.5 - Carpal maximum fiexion (above) and coffin maximum extension
(below) angles i SD for lame (gray shading) and contralateral (striped shading)
forelimbs of the lame horses (group 1) .............................................................. 53

Figure 4.2.1 - Vertical (above) and longitudinal (below) ground reaction force
traces for “mean lame” (solid gray line) and “mean contralateral" (gray line with
circles) forelimbs of the lame horses (group 1), and “mean sound” (solid black
line) forelimb of the control horses (group 2) ..................................................... 56

Figure 4.2.2 - Vertical and longitudinal ground reaction force traces for lame
(solid gray line) and contralateral (gray line with circles) forelimbs of the lame
horses (group 1) ................................................................................................. 58

Figure 4.2.3 - Braking impulse : SD for lame (gray shadings) and contralateral
(striped shadings) forelimbs of the lame horses (group 1) ................................. 60

Figure 4.4.1.1 - Kinegram of lame (dashed gray line) and contralateral (solid
black line) forelimbs of subject 1 (above) and subject 3 (below) ........................ 63

Figure 4.4.2.1 - Force vector diagrams of GRFs from lame and contralateral
forelimbs of the lame horses (group 1) .............................................................. 66

xi

CHAPTER 1

INTRODUCTION

Orthopedic injuries in racing Thoroughbred horses often involve severe
trauma to bones and soft tissues of the limb. These injuries have a severe,
sudden, and dramatic onset, and their presence limits future performances
(Cohen et al., 1997). In Thoroughbreds, 90% of musculoskeletal racing
injuries involve the forelimbs, and these injuries are primarily located in the
carpal joint, metacarpophalangeal joint and metacarpal bone. Most of the
catastrophic racing injuries, however, involve the suspensory apparatus of the
forelimbs (Peloso et al., 1994). In young horses, the suspensory ligament is
the weakest link in the suspensory apparatus. Active training appears to
strengthen the suspensory ligament, so in racing horses the weakest
component of the apparatus becomes the proximal sesamoid bones
(Bukowiecki et al., 1987). Moreover, during exercise, fatigue of the soft
tissues that support the distal limb may allow excessive extension of the
metacarpophalangeal joint, when tensile forces may exceed the
biomechanical tolerance of the structures, leading to failure of the bone or soft
tissues (Hubert et al., 2001 ).

Following an orthopedic injury, the main clinical Sign is lameness.

Lameness is defined as a clinical manifestation of abnormal gait, and it is

indicative of a structural or functional disorder in one or more limbs, or in the

back (Wittmann, 1931; West, 1984; Stashak, 1987; Wyn-Jones, 1988; Speirs,
1994; Wilson & Keegan, 1995). Under clinical conditions, the diagnosis of
lameness is based on experience and ability of the clinician to detect
asymmetries and abnormalities in the locomotion pattern. During the
lameness evaluation, the equine clinician assesses the severity of lameness
and assigns a grade based on a standard scoring system (Swanson, 1984)
from 1 (least severe) to 5 (most severe). This grading system offers
subjective information about the severity of lameness, but its use has been
found to be unreliable because of a lack of agreement between clinicians with
different levels of expertise when scoring a mild lameness (Keegan et al.,
1998). Further, the grading score system poorly correlates with objective

measures of locomotion, such as ground reaction forces (Fuller et al., 2002).

The biomechanical changes associated with lameness lead to
alterations in the limb loading profile, and in the force distribution within and
between the limbs (Goodship et al., 1983; Morris 8 Seeherman, 1987;
Merkens et al., 1988). For instance, lame horses show a progressive
reduction of the vertical and longitudinal braking components of the ground
reaction forces (GRF). Asymmetry of GRFs between lame and sound limbs
has been shown at walk (Merkens & Schamhardt, 1988) and trot (Morris &
Seeherman, 1987; Clayton et al., 2000). The peak vertical force (F2) of the
lame limb has been found to be 11.5% to 27% less than that of the sound
limb, depending on the severity of lameness (Morris & Seeherman, 1987;

Clayton et al., 2000). In addition, changes in limb kinematics, such as joint

angular patterns, can also indicate local pain or mechanical disturbances
associated with lameness (Stashak, 1996). For example, a significant
(p<0.05) reduction of approximately 6° in maximum metacarpophalangeal
(fetlock) joint extension angle, and significant reduction of approximately 38°
in maximum coffin joint fiexion angle have been reported during
experimentally induced forelimb lameness (Buchner et al., 1996). An
important feature of biomechanics is the movement variability, found to be
intrinsic in all movement patterns (Bernstein, 1967). The wide variety of
orthopedic diseases is likely to increase the biomechanical variability between
horses, which is supported by disease-specific changes found in GRFs
between horses (Williams et al., 1999). In addition to inter-individual variation,
a decrease in intra-individual variability appears to play an important role as a
mechanism to diminish pain in horses (orthopedic disease) (Peham et al.,

2001)

Specialized diagnostic techniques that are able to detect these
biomechanical changes provide a more sensitive quantitative assessment of
lameness severity but, so far, they have not been widely used to assist in
detection of lameness. Such techniques, as well as improving lameness
diagnostics, could be used to monitor progress and assess the success of
therapy (Keg et al., 1994; Clayton et al., 1998; Theyse et al., 2000).
Clinicians, using more sophisticated diagnostic tools, may more efficiently

address the level of discomfort in horses caused by lameness.

Biomechanical analysis offers a method to objectively assess
lameness and improve our understanding of gait abnormalities, adaptation to
lameness, and improvements in response to therapy. This study aims to
identify sensitive biomechanical variables to assist in the diagnosis of
lameness, and to develop innovative graphical displays to facilitate the
interpretation of biomechanical data. Furthermore, this study will investigate

the effects of lameness on the variability of movement.

CHAPTER 2

LITERATURE REVIEW

This chapter provides an overview of the history and development of
equine locomotion analysis as a potentially advanced objective measure of
locomotion. The epidemiology and mechanisms of musculoskeletal injury will be
described, including conventional methods of diagnosis of lameness and
potential applications of gait analysis in the diagnosis and evaluation of
locomotor abnormalities in horses. Existing graphical displays of biomechanical
data will be illustrated and, finally, biomechanical variability will be addressed in

relation to its potential value in lameness assessment.

2.1 HISTORY AND TECHNIQUES OF EQUINE GAIT ANALYSIS

In 1872, Eadweard Muybridge performed a series of photographic studies
of equine locomotion in Palo Alto, California, at a farm owned by Leland
Stanford, the founder of Stanford University. After a period of traveling in Mexico
and Central America, Muybridge returned to California and later continued his
photographic work at the University of Pennsylvania. He used 24 Single lens
cameras that were trigged in series to capture sequential pictures of the
locomotion of people, horses and other animals (Figure 2.1.1). In the meantime,

in France, the physiologist and professor Jules-Etienne Marey was investigating

equine gait analysis with different techniques. To discriminate between stance
and swing phases Marey invented an “exploratory shoe" that detected an
increase pressure in a small rubber balloon attached to the horse’s foot that was
designed for use on soft surfaces, while an “air-filled bracelet" was used on hard
surfaces. To measure vertical movements of the withers and croup, he adapted
two collapsible drums that were fastened to the withers and croup.

Figure 2.1.1 - Chronographic method of motion analysis in a galloping horse
entitled “The horse in motion" (Muybridge, 1887).

 

 

 

 

 

 

 

 

 

 

 

As the body falls and moves over the supporting foot, vertical, horizontal,

and rotatory forces are generated on the floor that can be measured with
appropriate instrumentation. The ground reaction forces (GRFs) are equal in
magnitude and opposite in direction to the forces exerted by the weight-bearing
limb. From knowledge of limb morphometrics, kinematics and ground reaction
forces, the stress imposed on the joints and the muscular torque needed to

control joint action can be calculated (Perry, 1992). In equine locomotion

research, quantification of forces related to locomotion were first measured using
pressure sensors attached to the shoe under the hoof and accelerometers
attached to the limbs to measure the hoof-ground contact durations at the
various gaits (Marey, 1873). Subsequently, strain gauge transducers were built
into shoes of draft horses to evaluate horizontal and vertical forces between the
hoof and ground (Bjorck, 1958). Vertical and horizontal components of forces
acting on the hoof have subsequently been measured in Standardbreds (Quddus
et al., 1978) and Thoroughbreds (Geary, 1975; Bartel et al., 1978). Today,
measurement of GRFs is also accomplished with force plates.

Force plates are the most common device used to measure ground
reaction forces and are fundamental for biomechanical analysis. Force plates
enable the determination of variables such as stance duration, magnitude of
vertical, horizontal and transverse forces, time for peak forces, impulses, and
center of pressure (Merkens, 1993; Shamhardt et al., 1993; Clayton, 1996;
Wilson et al., 2001). A force plate consists of a rigid platform with piezoelectric or
strain gauge transducers at the comers. By having three sensors set at right
angles (orthogonal) to each other, it is possible to measure the vertical load and
the horizontal shear forces in longitudinal and medic-lateral directions. Through
additional processing of these data, the related rotatory moments, center of
pressure, and ground reaction force vectors can be determined (Perry, 1992).

Biomechanical research in animal locomotion has undergone tremendous
technological innovations. For many years cinematography was the preferred

technique for detailed quantitative analysis of locomotion, but it has some major

limitations, including the long delay in obtaining results and the high cost of
purchasing and processing cine film. Videography overcomes these limitations,
but the early video systems had poor spacial resolution. Today, many measuring
techniques are highly developed allowing precise and timely quantification of
many aspects of animal movement. Automated and semi-automated gait
analysis systems, such as SELSPOT, CODA-3, VICON and Motion Analysis,
allow data to be collected fully automatically (Back at al., 1995). This modern
approach requires active (e.g. LEDs with SELSPOT), or passive (e.g. retro-
refiective with Motion Analysis) markers to be attached on the body to the skin
overtying anatomical landmarks on the bones. Opto-electronic or infrared
cameras track the markers with their location being recorded into a computer

(Figure 2.1.2).

Figure 2.1.2 — Gait analysis being performed using Motion Analysis System.

 

Although the automated biomechanical data collection systems provide
many benefits for equine locomotion research, there are also some drawbacks.
The high cost of purchasing and maintaining a gait analysis system and the
technical skills required to operate it generally limits its use to the laboratory
environment. The use of skin-based markers generates artifacts, especially at
the proximal joints, due to skin displacement over the Skeleton during locomotion
(van Weeren et al., 1990). Skin displacements can be as large as 12 cm, which
is sufficient to change the entire shape of the angle-time diagrams (Back at al.,
1994). For joints such as the shoulder and elbow, data with skin displacement
artifacts cannot be used for absolute angular computations or for measuring
muscle or tendon lengths based on limb kinematics (Clayton & Schamhardt,
2001). This skin displacement has been quantified and correction algorithms
have been developed for walking and trotting Dutch wannblood horses (van
Weeren et al., 1990; 1992). However, these correction algorithms are only valid
for horses with similar conformation, moving at the same gait, and with the same
speed. Despite conformational differences, these algorithms have been used in
other breeds, such as in Thoroughbreds, due to the lack of more accurate data.
Although the use of this algorithm in Thoroughbreds may not be ideal, it is better

than not using any correction.

2.2 APPLICATIONS OF EQUINE GAIT ANALYSIS

The use of domestic animals for multiple tasks, such as work, pleasure
and sport, emphasizes the necessity of selection methods for the fastest,
strongest and most beautiful animals (Schamhardt et al., 1993). Until recently,
selection was based on subjective criteria, which can lead to unfair judgments
(Leach, 1987; Ratzlaff, 1989). The need for more objective methods of evaluation
was one of the factors that led to the explosion of equine locomotion research,
that began in the 19703 and continues today (van Weeren, 2001). Studies
included computation of internal forces in the digit (Bartel et al., 1978); kinematic
differences between distal portions of the forelimbs and hindlimbs (Back at al.,
1995), effects of trotting speed on muscular activity and kinematics (Robert et al.,
2002); effects of morphological variation on biomechanical data (Lewis et al.,
2002); description of navicular bone movement in vitro (van Dixhoom et al.,
2002) and several others. Gait analysis uses biomechanical principles to make
objective measurements of locomotion and provide a detailed quantitative
description of movement patterns in humans and animals (Bartlett, 1997).

Biomechanics is defined as the science that examines forces acting upon
and within a biological structure and the effects produced by such forces (Hay,
1982). Vertebrate locomotion is controlled by mechanical principles that are
expressed in Newton’s three Laws of Motion (see Gray, 1968). The First Law
states that if the body of an animal is at rest relative to its environment, it can

only be set in motion by the application of an external force, and, consequently, if

10

an animal is to move its body by its own unaided efforts, it must elicit a force from
its external environment. The Second Law states that the sum of all forces (2F)
acting on a body of a given mass (m) equals the mass times acceleration (a) of
that body (F = ma). The Third Law states that for every action there must be and
equal an opposite reaction. Translated into biological terms, this can be
expressed by saying that, in order to subject its body to a fonlvard propulsive
force, the animal must simultaneously exert an exactly equal, but opposite,
backward force against its external environment (Gray, 1968).

Biomechanics uses two complementary approaches to study the body in
motion: kinematics and kinetics (Barrey, 1999). Kinematics is the study of the
geometry of motion, while kinetics is the study of internal (muscle activity) and
external forces (e.g. ground reaction forces, GRF) acting upon the body (Nigg,
1999). Kinematic data are expressed as temporal (timing), linear (distance or
displacement), and angular measurements that describe the movements of the
body segments and joint angles. Within the past few years, computerized
kinematic gait analysis has been used to study movement of clinically normal and
lame horses (Barrey, 1999). Several investigators have described various
kinematic variables in horses with normal gait patterns (Back et al., 1995;
Johnston et al., 1996; Back at al., 1996; Pourcelot et al., 1997; Clayton et al.,
1998), with superior movement qualities (Holmstrdm et al., 1994; Morales et al.,
1998), and with specific abnormalities (Buchner et al., 1995, 1996; Pourcelot et
al., 1997; Keegan et al., 1998; 1997). The joint movement patterns of the equine

limbs are important indicators of both physiologic locomotor capacity (Back at al.,

11

1994; Holmstrém et al., 1994) and gait disturbances due to lameness (Buchner
et al., 1996; Keegan et al., 1997). In 1994, Back et al. showed that some forelimb
kinematic variables such as stride and swing duration, scapular rotation, maximal
fetlock extension, and forelimb maximal retraction correlate well with judged
scores for gait quality in young trotting Warrnbloods. Head and trunk movements
have been described as sensitive indicators of lameness in horses with fore and
hind limb lameness (Peloso et al., 1993; Buchner et al., 1996). Furthermore,
fetlock extension, carpal flexlon and stride length have been shown to decrease
with induced carpal lameness (Back et al., 1993). In a model of supporting limb
lameness, in each pain is induced by pressure on the hoof sole (Buchner et al.,
1995), the most striking changes were found in the pattern of the fetlock joint,
where the maximal extension angle decreased with increasing lameness
(Buchner et al., 1996) (Figure 2.2.1), and in the coffin joint, where peak flexion in
the first half of stance was reduced with each degree of lameness.
Compensatory mechanisms in the contralateral limb were also found, where both

fetlock extension and coffin flexion were increased (Buchner et al., 1996).

12

Figure 2.2.1 - Fetlock joint angle pattern of lame (left) and contralateral (right)
forelimbs using a model in which lameness was induced by pressure on the hoof
sole. = lameness degree 0 (non-lame), _ _ _ = lameness degree 1,

- - - = lameness degree 2 (Buchner et al., 1996).

 

 

 

 

 

 

 

 

 

 

 

A - ‘5?
E 70 fetlock g
5 5° .5
‘3: 3° ,/ '3.
g 10 l 5
.10 ‘§_{‘ C _ y
a: '30 . - w . E -30 V - - r
0 20 4O 60 80 100 O 20 40 60 80

Time (% of stride) Tlme (“/o of stride)

Arrows indicate the events selected by Buchner et al. (1996) for quantitative
analysis

In contrast to the distal joints, the proximal joints are considered to play a
more active role in lameness management. The movement of proximal joints is
more dependent on muscular control than the distal joints, such as the fetlock,
where passive support by the interosseus (suspensory) ligament is the most
significant factor (Buchner, 2001). The shoulder normally flexes as the forelimb is
loaded, and interestingly in the lame forelimb, this flexion was increased as a

means of controlling the loading of the distal limb (Buchner et al., 1996) (Figure

2.2.2).

13

Figure 2.2.2 — Shoulder joint angle pattern of lame (left) and contralateral (right)
forelimbs. = lameness degree 0 (non-lame), _ _ _ = lameness degree 1,
- - - = lameness degree 2 (Buchner et al., 1996).

 

 

 

 

 

 

 

 

 

 

’u? - ’«7 -
3 10 shoulder 8 20 shoulder
a o “ "‘ g 10‘ -\
2 i ‘ _ g ‘
O) a . s _ v‘
\ A.

5 -10 g > 5 0 i;
8 x ‘ 8 j 15
'§ '§ «— v
‘1’ ‘20 L . - 9 ~10 a - -
E o 20 40 so so 100 U- o 20 4o . 60 so 100

Time (% of stride) Time (% of stride)

Arrows indicate the events selected by Buchner et al. (1996) for quantitative
analysis

In regard to GRFS, the vertical force represents the supporting function of
the limb. In sound horses, the vertical force has a peak magnitude of the order of
60% and 90% of the body mass at walk and trot, respectively (Clayton &
Schamhardt, 2001). However, recent studies have shown that the vertical force
in sound trotting horses can reach 110% of the body mass (Mullineaux &
Clayton, in review). During trotting, the vertical force trace shows a small peak
immediately after ground contact during the impact phase. The force trace then
rises smoothly to peak approximately at the middle part of the stance phase.

Subsequently, the vertical force decreases to lift off (Figure 2.2.3).

14

Figure 2.2.3 — Mean (n = 6) vertical ground reaction forces during the stance
phase for sound horses (mean value for 2 forelimbs) and lame horses (lame and
compensating forelimbs) with superficial digital flexor tendonitis (Clayton et al.,
2000)

 

  
 
     
      

 

 

 

14
U)
i: ,o' """"""""
Z 12‘ .......
8
x. ‘0.
.9
8
=8 "
(U
2 e-
E
3 - Sound
g, ‘H m Compensating
75 9” Lame
.2 2‘
t
0)
> of

o 20 do so no 100

Percentage of stance phase

The longitudinal force has a negative (braking) phase followed by a
positive (propulsive) phase (Figure 2.2.4). The peak longitudinal force represents
10 to 15% of the horse’s body mass at the walk and trot (Clayton & Schamhardt,

2001). Marked negative spiking occurs during the impact phase at the trot.

15

Figure 2.2.4 — Mean (n = 6) longitudinal ground reaction forces during the stance
phase for sound horses (mean value for 2 forelimbs) and lame horses (lame and
compensating forelimbs) with superficial digital flexor tendonitis (Clayton et al.,
2000)

 

 

 

 

 

3’ 15

\

E

G) 1.0 -

8

.9 ..........
c 0-5 " ‘ ‘Q.
.9

.H 0

3 col

22 . f v I ‘ 1

'O I’

g -o.s « fl '

9 i '. .' .. Sound
a: _1 o 4 | ,. C _
a . : ‘. 1" In. ompensatlng
2% ' ‘, ' coo Lame

:3 -1.5 4 ,‘ ‘ _

5': ' s ,' ‘..

O) o .0

8

_‘ .200

Percentage of stance phase

Despite the rapid progress of equine locomotion research and the
introduction of new concepts of gait analysis that enable the identification of gait
abnormalities, the interpretation of results may be complex and further hampered
by the large inter and intra-individual variability of subjects that leads to lack of
statistical significance (van den Bogert & Schamhardt, 1993). Moreover,
understanding of how the horse adapts to specific injuries in one or more limbs is
still limited. Due to the complexity in the interpretation of results, increased type II
error, and poor understanding of mechanisms of gait adaptation in response to
specific injuries and lameness, it is clear that the study of factors affecting

locomotion and lameness must be of high priority (Leach & Crawford, 1983).

16

2.3 MUSCULOSKELETAL INJURIES IN RACING THOROUGHBREDS

2.3.1 Epidemiological aspects

Musculoskeletal injuries are the most common condition afflicting
racehorses (Rossdale et al., 1985). In North American racing, the overall
incidence of musculoskeletal injuries ranges from 3.3 to 7.3 per 1000 starts,
depending on variables such as reporting criteria and degree of follow up (Hill,
2003). The rates for training injuries may be somewhat higher, although accurate
acquisition and evaluation of these data is difficult (Mundy, 1996). Early studies
demonstrated that lameness is the most common reason that horses in training
failed to race (Jeffcott et al., 1982; Rossdale et al., 1985), accounting for 67% of
days lost. As well as the discomfort and performance loss, lameness is estimated
to cost the USA horse owning public approximately 678 million dollars per year
(USDA, 2001).

The forelimbs are involved in 90% of the racing injuries (Peloso et al.,
1994), particularly in racing Thoroughbreds, with 95% of lameness occurring at
the level of or distal to the carpus (Ross, 2003). The high incidence of injuries in
the forelimb has been suggested to be due to the increased load on the forelimbs
observed at center and gallop, along with the cranial Shift in the center of gravity
promoted by the gait itself and the mass of the rider (Ross, 2003). Overall, the
left forelimb is most frequently involved, with injuries particularly occurring in the

turns. When raced in a counter-clockwise direction, horses are usually on the left

17

lead in the turns and on the right lead during the straightaway (Palmer, 1986). It
has been shown that the highest vertical force is on the lead forelimb at the
gallop (Ratzlaff et al., 1990), which increases the injury risk on the lead limb
(Peloso et al., 1994). Among the causes of lameness, 85.5% of all racing injuries
occur from the carpus to the metacarpophalangeal joint (Peloso et al., 1994),
with the metacapophalangeal joint alone representing 14% (Rossdale, 1985).
Multiple risk factors are associated with mild to catastrophic (fatal)
musculoskeletal injuries in racehorses. Increasing evidence suggests that pre-
existing pathologic conditions could play a role in the development of injuries
(Krook et al., 1988; Stover et al., 1992; Peloso et al., 1994; Cohen et al., 1997).
Horses with lameness grading score of 1 and abnormalities of the suspensory
ligament, diagnosed during pre-race physical inspections, were demonstrated to
have odds of injury to the suspensory apparatus of the forelimb 34 times greater
than horses with no abnormalities (Cohen et al., 1997). Additional risk factors
such as poor track design (Cheney et al., 1973; Hill et al., 1986; Clanton et al.,
1991) and variation in the number of days between races have been associated
with injury (Stover et al., 1992). Time interval between races of less than 3 weeks
has been observed to increase the occurrence of injury (Stover et al., 1992).
Moreover, a strong perception exists in the racing industry that turf courses are
safer than dirt courses. On the other hand, studies on the real influence of turf
racing versus dirt racing on the incidence of musculoskeletal injuries in

Thoroughbred racehorses have been criticized (Kobluck, 2003) based on a

18

personal judgment that considers most of the horses racing on turf higher quality

athletes, thus the data are not directly comparable.

2.3.2 Mechanisms of musculoskeletal injury

Musculoskeletal injuries occur as a result of biomechanical stresses of
abnormal intensity, duration, and frequency placed on bones, joints and
attachments of the skeleton during race training and racing (Cohen et al., 1997).
The main locations of injuries in the skeleton are hyaline cartilage, bone, and
fibrous tissues. There is increasing evidence that a significant proportion of
fractures affecting Thoroughbred racehorses involve fatigue processes. Fatigue
is a cumulative process. If the factors responsible for fatigue damage can be
identified, they may be manipulated before progressing to a catastrophic
conclusion (Riggs, 2002). A bone that is subjected to excessive cyclical loading
will undergo fatigue and its material properties will be progressively eroded. If the
rate of accumulation of damage is sufficiently rapid, the bone may be so
weakened that it becomes unable to withstand the normal loads of day-to-day life
(Riggs, 2002). If the intensity or frequency of loading is reduced, the bone may
show an adaptive response rather than failure.

Mild to moderate injuries, which are considered the most frequent during
races (Jeffcot et al., 1982), may progress to a more severe injury when the stress
overcomes the capacity of tissue repair (Estberg et al., 1996). This stage of the

pathology-injury Spectrum includes symptomatic lesions of joint capsule injury;

19

degenerative joint disease; chip, slab, lateral condylar, phalangeal, and proximal
sesamoid bone fractures (Figure 2.3.2.1); third carpal bone disease; bone

spavin; dorsal metacarpal disease; flexor tendonitis; and suspensory desmitis.

Figure 2.3.2.1 — Apical (left) and basilar (right) proximal sesamoid fracture
(White, 2002).

 

In contrast to mild to moderate injuries, catastrophic injuries are less
common. Catastrophic injuries often involve a failure of the suspensory
apparatus of the forelimbs (suspensory ligament and its branches, proximal
sesamoid bones, and distal sesamoidean ligaments) (Figure 2.3.2.2). During
exercise, fatigue of the proximal portion of the suspensory apparatus may allow
excessive extension of the metacarpophalangeal joint, producing tensile forces
that exceed the biomechanical tolerance of the supporting structures leading to

failure of the bone or soft tissues (Hubert et al., 2001) that compromise the

20

support of the metacarpophalangeal joint (Estberg et al., 1992; Jonhson et al.,
1994; Estberg et al., 1996; Kane et al., 1996). The weakest components of the
suspensory apparatus are those most likely to fail. The proximal sesamoid bones
are the weakest component of the suspensory apparatus in young racing horses,
when the suspensory ligament has been strengthened with exercise (Bukowiecki

et al., 1987).

Figure 2.3.2.2 — Suspensory breakdown injury with a comminuted fracture and
distraction of both sesamoid bones (Auer & Stick, 1999).

 

Racing injuries involving the fetlock joint may cause moderate to severe
lameness depending on the etiology. Non-fracture injuries caused by acute or
repetitive overload, such as synovitis, chronic proliferative (villonodular) synovitis,

osteoarthritis, subchondral bone injury, and sesamoiditis, are associated with

21

moderate lameness that may increase as the injury progresses (Stashak, 2002).
Articular fragments at the dorsal or palmar aspects of the proximal phalanx;
apical, abaxial, and basilar sesamoid fractures; and fragments from the sagittal
ridge of the MCI“ may cause a more pronounced and sudden lameness
(Stashak, 2002). Major articular fractures that often lead to euthanasia, including
sagittal and dorsal fractures; collateral avulsion injuries of the proximal phalanx,
mid-body; abaxial or basilar fragments of the proximal sesamoid bones; and
condylar fractures of MCI“ promote severe lameness either at the end of or

during a race (Stashak, 2002).

Videotape analysis of racing accidents suggests that injuries frequently
occur when the lead limb is changed, the jockey uses the whip, or the horse is
moving obliquely across the track (Ueda, 1991; Ueda et al., 1993). Stumbling
during race is associated with catastrophic, and career ending injuries (Cohen et
al., 1997). Physical interaction with another horse during the race is often
associated with catastrophic injuries and injury of the superficial digital flexor

tendon of the forelimb (Cohen et al., 1997).

2.4 CLINICAL EVALUATION OF LAMENESS

Horses manifest locomotion abnormality by lameness, which may indicate

structural or functional disorder in one or more limbs or in the back (Seeherman,

1999). This condition can be caused by trauma, congenital or acquired

22

abnormalities, infection, metabolic disturbances, circulatory and nervous
disorders, and any combination of these (Stashak, 2002).

Experience, acquired by years of clinical practice, working and learning
from experienced practitioners, is required for accurate lameness evaluations
(Ross et al., 2003). More than that, the evaluation of lameness requires a
detailed knowledge of anatomy, an understanding of kinematics, and an
appreciation for geometric design and resultant forces (Stashak, 2002). The
development of the skills required to become a true lameness diagnostician
requires a thorough, somewhat methodical approach (Ross et al., 2003). Equine
practitioners carefully evaluate lameness following four main steps: (1)
anamnesis; (2) identification of the lame limb or limbs through observation at rest
and exercise; (3) palpation and manipulation of the lame limb or limbs to allow
identification of the Site of pain; and potentially (4) diagnostic anesthesia and
imaging diagnostics to clarify the location of pain, the nature of the problem, and
the extent of injury. The severity of lameness is categorized as mild, moderate, or
severe, or by using a qualitative and semi-quantitative grading score system (e.g.

Swanson, 1984; Table 2.4.1).

23

Table 2.4.1 — Lameness grading score system (Swanson, 1984).

 

GRADE LAMENESS DESCRIPTION

 

 

 

 

 

 

0 Lameness not perceptible under any circumstances
Lameness difficult to observe; not consistently apparent regardless
1 of circumstances (e.g., weight carrying, circling, inclines, hard
surface)
Lameness difficult to observe at a walk or trot in a straight line;
2 consistently apparent under some circumstances (e.g., weight
carrying, circling, incline, hard surface)
3 Lameness consistently observable at a trot under all circumstances
4 Lameness obvious; marked nodding, hitching, and/or shortened
stride
5 Lameness obvious; minimal weight bearing in motion or rest;

inability to move

 

types: (1) supporting limb lameness is evident when the foot first contacts the
ground or when the limb is supporting weight (stance phase). It is often caused
by injury to bones, joints, soft tissue (e.g., ligaments and flexor tendons), motor
nerves, and/or the foot. (2) Swinging limb lameness is seen when the limb is in
motion and it is often caused by pathologic changes involving joints, muscles,
tendons (primarily extensors), tendon sheaths, or bursae. (3) Mixed lameness is
seen both when the limb is moving (swing phase) and when it is supporting
weight (stance phase). It may involve any combination of the structures affected

in swinging or supporting limb lameness. (4) Complementary or compensatory

In addition to the severity, lameness is usually classified into one of four

24

lameness occurs when pain in one limb causes uneven distribution of weight on
another limb or limbs. This can produce lameness in a previously sound limb. For
instance, lameness of the right forelimb may promote compensatory lameness in
the contralateral forelimb and/or ipsilateral hindlimb (Stashak, 2002).
Identification of such types and severities of lameness requires an ability
to recognize the subtle as well as the obvious signs, and to possess the
knowledge to interpret the observations. The most convenient sensor is the
trained eye of the practicing clinician, that permits assessment of the problem at
any time and in any environment (Perry, 1992). Visual observations, along with
the lameness score system, provide semi-quantitative information with good
repeatability among experienced clinicians (Back at al., 1993). However, there is
considerable variation between clinicians with different levels of expertise
(Keegan et al., 1998) and poor correlation between the scores assessed by
visual observations versus objective measures (e.g. GRFS) (Fuller et al., 2002).
The variability and inconsistency of subjective lameness evaluations justifies the
need for development of objective ways to improve the evaluation of lameness in

horses.

2.5 USE OF BIOMECHANICS IN LAMENESS EVALUATIONS

Some types of lameness cause pain during loading so the horse tries to

minimize this pain by changing various aspects of the locomotion pattern

(Buchner, 2001). These changes are manifested by asymmetries in temporal and

25

spatial kinematics, as well as force distribution variables between the limbs
(Clayton, 1987; 1988; Girtler, 1988). Biomechanical or gait analysis can detect
these changes in the locomotion pattern. Variables such as stride length, stride
duration, stance duration, peak vertical and longitudinal ground reaction forces,
and joint angular kinematics can be used to detect gait abnormalities (Leach,
1983; Back at al., 1993; Schamhardt et al., 1993; Buchner et al., 1996; Barrey,
1999; Peham et al., 2001). However, disease specific changes in the locomotion
pattern and mechanisms of lameness adaptation are still poorly described.

In some areas there are controversial findings, for example, in the effects
of supporting limb lameness on temporal stride variables. Some researchers
have described a shortening of the stance phase after induction of carpal
lameness as a mechanism to diminish pain (Ratzlaff et al., 1982), while others
described a lengthening of the stance phase using the same model of lameness
(Morris & Seeherman, 1987) or when lameness was induced by pressure on the
hoof sole (Galisteo et al., 1997), with both limbs kept on the ground longer
(Buchner et al., 1995; Keegan et al., 1997; Back et al., 1993; Galisteo et al.,
1997). In contrast to obvious supporting limb lameness, subclinical lamenesses
do not Show significant temporal deviations from the sound stride pattern
(BuchneretaL,1995)

Lameness can also be detected by changes in the ground reaction forces
(Tietje, 1992; Buchner et al., 1995). A decrease in both vertical (see section 2.3,
Figure 2.2.3) and longitudinal (see section 2.3, Figure 2.2.4) ground reaction

forces in the lame limb is observed, while a compensation of amplitudes of the

26

non-affected limbs occurs (Gingerich et al., 1979; Silver et al., 1983; Merkens &
Schamhardt, 1985, 1988). The lengthening in stance phase duration allows the
impulse to be maintained with a lowest force amplitude (Morris & Seeherman,
1987). The decrease in braking horizontal force is suggested to be due to an
ineffective deceleration of the limb (Morris & Seeherman, 1987).

Joint angle patterns are also influenced by lameness. Supporting limb
lameness cause striking changes in the joint angle patterns especially at the
distal joints such as the fetlock (see section 2.2, Figure 2.2.1) and coffin joints
(Buchner et al., 1996). Changes in joint angles usually correspond to the
decrease in vertical GRF observed in the lame limb (Riemersma et al., 1988; see
section 2.2, Figure 2.2.3). Compensatory changes may be seen in the
contralateral limb (Merkens & Schamhardt, 1988; Morris & Seeherman, 1987).
Limb retraction and protraction angles also change with lameness to adjust to the
orthopedic pain. After sole pressure-induced lameness, maximum protraction of
the forelimb is greater than before the induction of lameness (Buchner et al.,
1996; Keegan et al., 2000), while both lame and contralateral forelimbs are less
retracted at the end of the stance phase (Buchner et al., 1996).

Advances in the equine biomechanics field have enabled detailed studies
of experimentally induced lameness (Morris & Seeherrnan 1987; Merkens &
Schamhardt 1988; Back at al 1993; Peloso et al. 1993; Buchner et al. 1995,
1996; Deuel et al. 1995; Clayton et al., 2000) or naturally occurring lameness
(Clayton, 1986; Williams, 2001; Peham et al., 2001; Clayton et al., 2002). Gait

analysis provides a method of quantifying the horse’s locomotion abnormality

27

with reliable instrumentation and permanent record of fact. Consequently, the
imprecision of subjective information is avoided, rapid and subtle events are
captured, and printed records of the horse’s motion pattern offer a reference
base for interpreting improvement. In order to be more easily applicable, gait

measuring techniques and the display of results should be simplified.

2.6 GRAPHICAL PRESENTATIONS IN GAIT ANALYSIS

Gait analysis produces voluminous numerical data that can ovenivhelm the
investigator with numbers that are difficult to interpret. Therefore, numerical data
are only useful after analysis in a manner that improves understanding and
knowledge (Schamhardt et al., 1993). One way to achieve a better level of
understanding of gait analysis is through the use of graphs or diagrams.

Basic locomotion information, such as the magnitude of the peak joint
angles (flexion and extension), can be easily extracted from biomechanical data.
However, the magnitude of joint motion as an independent item of information
may not be sufficient to identify the gait abnormality, as the timing of motion in
adjacent joints may be a critical factor (Peny, 1992). Graphical representations
such as variable-variable plots can maximize information about the movement,
suggest a specific gait abnormality, and facilitate understanding of the movement
(Mullineaux et al., 2000). For instance, joint angle-time diagrams (see section
2.2, Figures 2.2.1 and 2.2.2) can be used to describe coordination between joints

of the equine forelimb, and they can be related to stick diagrams (Figure 2.6.1),

28

which will improve the understanding of the movement the entire limb (Back at
al., 1995). Stick diagrams are used to illustrate the motion of the limb during the
stride using lines to represent the limb segments (Barrey, 1999). Overall, stick
diagrams facilitate recognition of left-to-right symmetry (or asymmetry) of
movement in humans (Perry, 1992) and animals (Pourcelot et al., 1997). Angle-
angle diagrams, for instance, carpal-fetlock diagrams, have also been found to
be useful in detecting forelimb coordination problems (Back et al., 1995).

Figure 2.6.1 — Stick diagram of the forelimb of an individual horse trotting at a

velocity of 4 m/s on a treadmill from left to right during a complete stride (Back at
al. 1995).

 

 

 

 

 

Force data output is expressed in terms of forces in three mutually
perpendicular planes, referred to as medic-lateral or transverse (Fx),
craniocaudal or longitudinal (Fy) and vertical (F2). The three force components
are usually displayed separately as force-time diagrams (see section 2.2.,
Figures 2.2.3. and 2.2.4). Although used widely in biomechanics, these types of

diagrams are difficult to interpret for pathological conditions because they give

29

little indication of the effects that abnormalities have on the alignment and
proximity of the force vectors relative to the joints of the distal limb, and the
turning effects (torques) that they generate around those joints (Stallard, 2000).
To permit a more immediately observable indication of the effects of these forces
in a human or animal, methods of displaying and analyzing force vectors have
been developed. A force vector is a line representing the magnitude and direction
of a force. Although forces are three-dimensional, the vector may be shown in
one plane (sagittal, frontal or transverse). In humans, force vector diagrams in a
sagittal plane, also called butterfly diagrams due to their shape, are used in gait
analysis to illustrate gait patterns in pathological conditions such as ostearthritis
and hip disorders, and also to assess intrinsic and extrinsic variability of GRF
within individuals (Rozendal et al., 1985; Khodadadeh, 1988; Baumann et al.,
1992; Rabufetti & Frigo, 2001). In horses, they have been used to illustrate GRF
patterns of Dutch Wannbloods at normal walk (Merkens et al., 1985), and also to
describe changes in GRF during different degrees of experimentally induced
lameness at walk (Merkens & Schamhardt, 1988) (Figure 2.6.2). The use of
these diagrams at trot for lameness diagnosis has not been described and their

effectiveness for this purpose needs further investigation.

30

Figure 2.6.2 — Vector diagram (or vectordynamogram) of mean (n = 4) Fy-Fz
force-force curves of all limbs during control section (HO) and moderate left
forelimb lameness (F3) (Merkens & Schamhardt, 1988).

 

 

 

 

 

 

 

FZ (N/kg) FZ (N/kg)
8 [ H0 81'-
6 6 ..
r-
4 L ' 4 II-
2 ~- 2 ~
I... l J 1 l 1 1 , __g__ޢ L 4
-1. -.5 0.0 -.5 1.0 1- -1.0 -.5 0.0 -.5 1.0 1.5
FY (N/kg) FY (N/kg)
* left IOTEIImb
right forelimb
——————— left hindlimb
------ right hindlimb

The use of sophisticated diagnostic tools, such as gait analysis, offers a
means of quantifying lameness in horses. However, to be considered an auxiliary
diagnostic tool, the resultant information needs to be illustrated effectively, thus
providing clear evidence and better understanding of gait abnormalities,
adaptation to lameness, and improvements through therapy. Graphical
representations are a valuable tool to be used in gait analysis studies, especially
to facilitate interpretation by clinicians who are not familiar with the analytical
method. It is timely to focus equine locomotion studies on the development of
more friendly ways to display numerical gait analysis data that will facilitate its

use in a clinical setting.

31

2.7 VARIABILITY OF MOVEMENT

The movement of healthy humans is characterized by slight, but
continuous, variation of the motion pattern (Yamada, 1995; Whitall & Getchell,
1996; van Emmerik 8 Wagenaar, 1996). This variation is an inherent component
of movement both within (intra-subject) and between (inter-subject) subjects
(Newell & Corcos, 1993). Every movement pattern is constrained by peripheral
sources of variation such as morphological, environmental and mechanical
(Bernstein, 1967; Higgins, 1977). Equine movement is also subjected to these
sources of variation. Although a small variation coefficient (2%) of kinematic
parameters is described within horses, the variation between horses is 2 to 3
times greater (Drevemo et al., 1980). Sound horses are considered to have
stable biomechanical variables, so the analysis of a relatively small number of
strides becomes representative of the gait pattern. Due to this gait repeatability, it
has been suggested that the analysis of 3 to 5 strides per horse is sufficient and
representative of both kinematic (Drevemo et al., 1980) and GRF (Schamhardt,
1996) patterns. However, in lame horses, the variability phenomenon has been
poorly described.

Variability is suggested to have detrimental effects on biomechanical
research. For instance, within and between subjects variation may affect the
reliability of individual scores and Significantly affect the statistical power of an
experiment (Bates et al, 1996; Dufek et al., 1995). The addition of trials can be

used to control within subject variability, or to control between subject variability

32

when subjects perform the same task in a similar manner (Dufek et al., 1995). A
more difficult situation arises when the source of variation is due to individuals
using different solutions (strategies) to accomplish the same task. A strategy is
defined as a selected neuromusculoskeletal solution for the performance of a
task. Humans select a strategy based on previous experiences, their
perceptions, and the resulting expectations (Dufek et al., 1995). Considerable
experimental evidence in support of individual strategies can be found in the
research literature (Bates et al., 1979; Loslever, 1993; Reinschmidt 8 Nigg,
1995). Performance differences (between subject variability) resulting from
different strategies threaten the external validity of a group design and often lead
to false support for the null hypothesis (Bates, 1989). Given these concerns, a
possible approach for avoiding these potential problems is to combine group and
single-subject designs to gain additional insight into the research problem of
interest (Dufek et al., 1995).

Single-subject, single case, and “n = 1” are all names for experimental
designs that involve one subject observed over time, while some variable or
factor is manipulated. The concept of single-subject analyses is not new.
Historically, the intensive study of individual human behavior in the areas of
psychology, psychiatry, and physiology began in the mid 1980’s (Barlow 8
Herson, 1984). The methodology often consisted of making repeated response
measurements to different stimuli. The results from these types of individual
studies produced important findings that, with adequate replication, often

provided indications for general results (Bates, 1996). As mentioned by Bates

33

(1996), the basic rationale for single-subject evaluation is simple: individuals are
unique, no two individuals are alike. Overall, this type of analysis intends to
overcome the effects of the individual selection of strategies to perform the same
task. Variability between horses affects the response to certain interferences,
such as drug treatment and shoeing, which differ qualitatively and quantitatively
in different animals (Clayton 8 Schamhardt, 2001). Impressive libraries of
statistical routines have been developed to extract trends in the data, to detect
differences between groups, or to identify a statistical significant response to a
certain treatment. In addition, the majority of equine locomotion studies are
based on a rather small number of subjects, which may be insufficient to give the
required power for a statistical analysis (Clayton 8 Schamhardt, 2001). The
individual locomotion responses, along with the limited number of subjects
available in equine biomechanical studies, emphasize the rationale for the
implementation of single-subject analysis as an additional analytical method.
More recently, in the area of human movement science, there has been a
resurging interest in the individual and, therefore, within individual variability. It
has been suggested that variability has a functional role to play in human
movement (Wheat et al., 2002) and it is fundamental for changes in the
coordination between body segments (Kelso, 1995). In addition, variability was
suggested to play a role in lower extremity coordination that attempts to
attenuate the large impact Shocks present during the stance phase of running,
thus playing also an important role in the prevention of injury (Heiderscheit et al.,

1999). In 1994, Buchner et al. used variability to characterize the stability of

34

motion as a parameter for habituation of horses on the treadmill. All horses in his
study showed within subject variation in stride variables. Variability has also been
used to determine the optimum speed for lameness evaluation on the treadmill
(Drevemo et al., 1980; Peham et al., 1998) and as an indicator of orthopedic pain
(Peham et al., 2001). In contrast to the high variability observed in running
humans (Heiderscheit et al., 1999), horses with orthopedic pain are described to
have low stride variability (Peham et al., 2001). Peham et al. suggested that this
low variability could be associated with the optimization of compensatory
mechanisms to reduce pain, and concluded that stride variability is an individual
parameter of lameness. This study represents a starting point for further
investigation of the function of variability in equine locomotion in lame and sound

conditions.

35

RESEARCH OBJECTIVES

1. To identify sensitive biomechanical variables to be used in the diagnosis
of fetlock joint lameness

2. To investigate variability in gait measures of Thoroughbred horses with
fetlock injury and to compare the variability of sound subjects at the trot

3. To explore the use of graphical representations of biomechanical data for

lameness detection

HYPOTHESES

H1 : Biomechanical variables are sensitive to detect lameness in Thoroughbred

horses with fetlock joint injury.

H2: Lame horses have less movement variability than sound horses.

36

CHAPTER 3

METHODS OF INVESTIGATION

3.1 SUBJECTS

Five lame Thoroughbred horses (group 1), with a history of career ending
unilateral racing injury in the metacarpophalangeal (fetlock) joint of the forelimb,
were selected for inclusion in this study (Table 3.1). An experienced clinician
performed lameness evaluation and attributed a standard lameness score
(Stashak, 1996) for each subject. Radiological study of the affected
metacarpophalangeal joint area was performed and the final diagnosis was
determined (Table 3.1).

Five sound horses were subjected to the same lameness evaluation
protocol and included as a control group (group 2) (Table 3.2). All horses were
used with approval of the Michigan State University All University Committee on

Animal Use and Care (AUF#07/01-113-00).

37

Table 3.1 — Descriptive characteristics and lameness diagnosis of subjects from

group 1 (R = right; L = left).

 

 

 

 

 

Sub'ect Age Mass Height Injured Lameness Lameness
’ (years) (kg) (m) forelimb score diagnosis
Basilar fracture of the
1 3 456 1.55 L 3 lateral proximal
sesamoid bone
Basilar fracture of the
2 4 484 1.61 R 2 medial proximal
sesamoid bone
Apical fracture of the
3 4 430 1.58 R 2 lateral proximal
sesamoid bone
Severe degenerative
4 6 517 1.66 L 2 jount disease wrth
exostosrs around the
fetlock joint
Degenerative joint
5 4 475 1.60 R 2 disease of the fetlock
joint
Table 3.2 — Descriptive characteristics of subjects from group 2.
Subject Breed Age (years) Mass (kg) Height (m)
1 Thoroughbred 18 508 1 .56
2 Thoroughbred 13 505 1 .6
3 Thoroughbred 15 537 1 .58
4 Warmblood 6 586 1 .6
5 Warmblood 6 518 1 .56

 

38

3.2 DATA COLLECTION PROTOCOL

The testing area consisted of a runway covered with high—density rubber
matting, into which a 60 cm x 120 cm2 force plate (AMTI LG6, AMTI, Watertown,
MA) was embedded. In addition to the force plate, the collection area was
equipped with an infrared gait analysis system consisting of 6 Falcon cameras
(Motion Analysis Corporation, Santa Rosa, CA, USA) and a capture volume of
5 x 2 x 3 m. The equipment allowed three-dimensional ground reaction force
(GRF) and kinematic data to be collected simultaneously throughout one
complete stride (hoof contact to hoof contact) as the horse moved along the
runway.

Calibration of the system was performed using a ‘seed and wand’
calibration method (Eva RT 3.2 User’s Manual Motion Analysis Corporation,
2002). The wand was 1 m in length with three retro-reflective, spherical markers
(38 mm diameter). The calibration process accounted for any geometric
distortion the camera lenses might have throughout the entire capture volume.
The location of the force plate was obtained by placing retro-reflective spherical
markers at each corner of the force plate. The three-dimensional component of
the GRF was adjusted to zero and the sensitivity matrix was adjusted to account
for “cross-talk" between the transducers, thus improving the force measurement
accuracy. The coordinate system used to measure GRF and kinematic data

consisted of transverse (X), longitudinal (Y) and vertical (2) axes (Figure 3.2.1).

39

The positive directions were upward for the vertical force, cranially for the
longitudinal force, and laterally for the transverse force.

Figure 3.2.1 — Coordinate system used during GRF and kinematic data
collection.

  
 

Force Plate

Z

Retro-reflective, spherical markers (25 mm) were attached on the lateral
aspect of the forelimbs overlaying the following landmarks: attachment of lateral
collateral ligament of the elbow joint on distal humerus (1 = elbow), distal edge of
ulnar carpal bone midway between lateral styloid process of radius and proximal
third metacarpus (2 = carpus), distal end of third metacarpus at attachment of
lateral collateral ligament of metacarpophalangeal joint (3 = fetlock), hoof wall at
the heel (4 = heel) and hoof wall at the toe (5 = toe). These markers allowed
further determination of carpus, fetlock and coffin joints flexion and extension
angles.

Markers 2 and 3 represented the approximate centers of the rotation of

carpal and metacarpophalangeal joints in the sagittal plane. The flexion and

40

extension angles were measured at the intersections of the segment lines at the
center of rotation of each forelimb joint. The location of the distal interphalangeal
(coffin) joint center of rotation was determined radiographically (see session 3.3).

Five successful trials were collected for each limb separately, in which the
forelimb being analyzed was fully on the force plate throughout its stance phase
with no other limb on the force plate at the same time. Velocity was normalized to
the horse’s body size to allow comparisons between horses of different sizes.
Height at the withers, which is proportional to the segment lengths of the
forelimbs and hindlimbs, was applied to scale the gait variables (Khumsap et al.,
2002). Overall, the horses’ trotting velocity was within a range of 3.0 to 3.3 ms".
GRF and kinematic data were collected at frequencies of 1200 Hz and 120 Hz
and filtered using 40 Hz and 15 Hz ButtenNorth low-pass filters, respectively
(Clayton et al., 2000). The GRFs were synchronized with the kinematic data.
Subsequently, the data were normalized to the horse’s body mass and to stride
duration to facilitate further comparisons between horses.

Correction algorithms were not applied to the kinematic data to account for
skin displacement because skin displacement for markers distal to the humerus
are small enough to be ignored (e.g. carpal and fetlock joints) (van Weeren et al.,

1988:1992)

41

3.3 DATA ANALYSES

Following data collection, 5 mm metal spheres were attached to the hoof
wall at the exact location of heel and toe markers. Latero-medial radiographs of
the hoof were taken from right and left forelimbs and later scanned to a
computer. The radiographs with the two reference markers at the heel and toe
were loaded into custom software (Radcof, Mary Anne McPhail Equine
Performance Center, East Lansing, MI) that determined the approximate location
of the center of rotation of the coffin joint. Two chords were drawn across the
articular surface of the coffin bone. The software constructed perpendicular lines
from the middle of each chord towards the center of the middle phalanx. The
intersection of these lines represented the center of rotation of the coffin joint
(Figure 3.3.1). The location of the center of rotation of the coffin joint was then
translated into a coordinate system (x, y) and located relative to the position of

the heel and toe markers in each frame throughout the stride.

42

Figure 3.3.1 — Latero-medial radiograph of the hoof showing schematic
determination of the coffin joint center of rotation.

  
 

Coffin bone

, . ' Toe marker
Heel marker _ /

Line representing a
coffin bone
segment

GRF and kinematic data for the forelimbs were analyzed using a custom
code written in Matlab® 6.1 (The MathWorks Inc, Natick, MA, USA). The code
accounted for time normalization to 101 data points from hoof contact to hoof
contact using cubic spline interpolation, and location of the coffin joint center of
rotation for each frame (coordinates extracted from Radcof software).

Joint angle measurements were calculated for each trial in a sagittal plane
on the anatomical flexor side of the forelimb joints. Flexion and extension were

defined as a decrease and increase in the angle value, respectively. Joint angle

43

traces, maximum extension and maximum flexion were obtained through angle-
time diagrams, and their time of occurrence was expressed as percent of stride
duration. Stride variables such as stride duration (seconds), stride length
(meters) and stance duration (percentage of stride) were also determined.
Kinegrams, an adaptation of stick figures, were used as graphical displays for
kinematic data. Kinegrams illustrate the locomotion pattern, and may assist in the
detection of asymmetries between the forelimbs. Changes in variability within the
kinematic variables were also investigated using standard deviation.

Vertical, braking and propulsive GRF traces were primarily obtained
through the use of force-time diagrams. Peak vertical, braking and propulsive
GRFs were determined and their time of occurrence was expressed as
percentage of stance duration. Vertical, braking and propulsive impulses were
calculated by time integration of the force curves. GRF peaks (N) and impulses
(Ns) were normalized to body mass where they were expressed as N/kg and
Ns/kg, respectively. In addition to force-time diagrams, a novel graphical display
was used in an attempt to improve the recognition of the lame forelimb. Force
vector diagrams were used to represent ground reaction forces in a method
adapted from humans (Rozendal et al., 1985; Khodadadeh, 1988) and horses at
walk (Merkens et al., 1985). These vector diagrams were applied in horses at trot
for the diagnosis of lameness. Changes in variability of GRF variables were also

investigated using standard deviation values.

44

3.3.1 Statistical analyses

The data were analyzed using two different statistical procedures, group
mean differences, and single-subject analysis. The mean value for each variable
was determined by averaging the 5 successful trials recorded for each subject.
Subject mean values were averaged to calculate mean values. Subsequently,
group mean differences within group 1 (“mean Iame” versus “mean
contralateral”) and between groups 1 and 2 (“mean lame" versus “mean sound”;
“mean contralateral” versus “mean sound”) were determined using dependent t-
tests. Single-subject analysis using dependent t-tests was performed for each
variable in the study, comparing lame and contralateral forelimbs with the data
gathered from the 5 trials for each forelimb. Dependent t-tests were used for the
analyses, because the data for right and left forelimbs were collected during
different trials, so they were not paired. The rationale for the use of single-subject
analysis arises from the high variability between subjects, which threatens the
reliability of the group statistical analysis by increasing type II error.

In addition, variability associated with the kinematic and GRF variables
was also analyzed. Mean differences in standard deviation between “SD lame”
and “SD contralateral” within group 1 were determined using paired t-tests. The
same procedure was performed to determine mean differences in standard
deviation between groups 1 and 2 (“SD lame” versus “SD sound", “SD

contralateral” versus “SD sound”).

45

The type I error was set at 5% (p<0.05) for all statistical analyses
performed. The assumption of normality was met in the majority of variables for
lame (lame and contralateral forelimbs) and control (sound forelimbs) groups.
Non-normally distributed data were only found in stance duration for lame
forelimb, propulsive impulse for contralateral forelimb and stride length for lame
and contralateral forelimbs (Shapiro-Wilk > 0.05). Only one variable, stance
duration between lame and contralateral forelimbs, was found to violate
homogeneity of variance (Levene’s test > 0.05). Despite these few violations,
inspection of the means and standard deviations supported the results of the
independent t-tests. Consequently, the normality and homogeneity of variance

were considered not to deter accurate inferential statistical analyses.

46

CHAPTER 4
RESULTS

In the angle-time and force-time diagrams in this chapter, sound, lame and
contralateral conditions will be displayed as lines in solid black, solid gray and

gray with circles, as follows:
o- Contralateral
~ Lame

— Sound

Kinegrams will have the lame and contralateral forelimbs displayed as

dashed gray and solid black lines, respectively:

Contralateral Lame

 

’——-——--

\ \.

4.1. KINEMATICS

Differences in joint angles between groups were restricted to maximum
extension of the fetlock joint (Table 4.1.1). Significant differences (p<0.05) within
group 1 (“mean lame” versus “mean contralateral” forelimb) were not found,
however, significant differences (p<0.05) between means of group 1 and group 2

were observed, where the maximum extension of the fetlock joint of the “mean

47

lame” was significantly smaller (-19°) than the “mean sound” forelimb (Figure
4.1.1).

Table 4.1.1 — Mean (SD) joint angles for lame (L) and contralateral (C) forelimbs
of the lame horses (group 1), and for the “mean sound” forelimb of the control
horses (S; group 2).

 

Joint Subject Maximum extension (degrees)§Maximum flexion (degrees)

 

 

 

 

 

 

 

L c L c
Carpus 1 V A. A I ‘8 183(07):: ‘. A, ._ 1871 15) ‘. I if 105(22): 114(14)
2 186(O.5)* 187(0.6) I 117(1.7)* 114(1.4)
3 185(0.7)* 184(0.5) 109(1.2) 108(1.7)
4 194(11):: 192(1.2) 114(5.8)* 123(1.3)
5 188(0.4) 188(0.2) 115(11): 123(1.3)
Group1 187(4.2) 188(2.9) 112(4.9) 116(6.5)
Group 2 (S) 187(4.1) 114(53)
Fetloékm1 I A if 221 (1 .7 )1: """"" 236(65)” ”143(80)” W1511V(“2li.0) M
2 217(0.4) 217(04) 140(2.8) 139(90)
3 237(04):: 240(03) 150(27) 155(6.7)
4 207(07):: 233(1.1) 160(1.4) 159(1.7)
5 218(1.4)* 234(10) 146(1.2)* 158(3.1)
‘ Group1 220(10.8)° 232(8.8) I 148(7.8) 152(8.1)
Group 2 (S) 239(6.0) 147(4.3)
Cofflnw1 H 240(1.7)* 225(517) 177(89) 168(16)
2 228(1.1)* 224(15) 171(07) 170(101)
3 219(06):: 231(1.4) 165(1.3) 163(3.2)
4 226(1.7)* 212(1.2) 187(33):: 171(1.3)
5 224(15): 198(3.5) 164(3.7)* 150(3.7)
G"?66p1“””227(7.8)““ 218(13.1) 173(9.5) 164(8.6)
Group 2 (S) 221(4.8) 172(29)

 

* Significant Single subject difference between L and C forelimbs (p<0.05)
3 Significant group mean difference between L and S forelimbs (p<0.05)

48

Figure 4.1.1 — Fetlock joint angle for the “mean Iame” (solid gray line) and “mean
contralateral” (gray line with circles) forelimbs of the lame horses (group 1), and
for the “mean sound” (solid black line) forelimb of control horses (group 2).

 

 

 

 

 

260 \ ~ 7
:1: E
A 220 - _
8
,5; r‘
e 180 7
2 .
o: .1
c l
< 140 ~ )
Stance Swing I
1 00 T T T I I:
O 20 40 60 80 100
Time (% Stride)

Arrows indicate the events selected for analysis (maximum extension in stance
phase, and maximum flexion in swing phase)

* Significant group difference between “mean lame” and “mean sound” forelimbs
(p<0.05)

Single-subject analysis demonstrated that the lame forelimb of the
majority of the subjects (horses 1, 3, 4 and 5) had Significantly (p<0.05) smaller
maximum extension of the fetlock joint than the contralateral limb, whereas in
subject 2 there was little difference between limbs (Figure 4.1.2). The largest
reduction in fetlock extension was seen in subject 4, which also had mechanical
limitation of joint mobility, which resulted in 207° of maximum fetlock extension in
the lame limb against 233° in the contralateral forelimb. The angle-time diagram
(Figure 4.1.2) shows this remarkable reduction in the range of motion of the

fetlock joint throughout the stride, where the joint trace of the lame limb is

flattened when compared to the contralateral and sound forelimbs. In cases of
49

moderate degenerative joint disease (e.g. subject 5) and proximal sesamoid

fracture (subjects 1 and 3), the reduction in fetlock extension is also evident but

not as prominent. Subject 2 did not demonstrate major differences between lame

and contralateral limbs, but both limbs had reduced maximum extension of the

fetlock joint.

Figure 4.1.2 — Fetlock joint angles for lame (solid gray line) and contralateral
(gray line with circles) forelimbs of the lame horses (group 1).

Angle (Degrees)

 

 

 

 

220 — m

 

 

 

 

 

 

260 _. - ~
220 T 0“; i’ 'o..\ I
180 3 °-., ,q/ 180 ~
140 — Z 140 ~
100 r r r T T 100
0 20 4o 60 80 100 0
260 1 ~ ~ - 260 ~-
220 I..‘~/_\.\ : 220 J . .
130 1 “\‘1 .~-'”’-“ 180 1"
140 4 ”i L 140 1
100 I I I I T 100
0 20 40 60 80 100 0
220 — /\ .
u" \. ’
180 1 ' -,/\_/
140 J ' '
100 r . r 4 T
o 20 4o 60 80 100
Time (%Stride)

50

No group mean differences within group 1 or between groups 1 and 2
were observed at the carpal and coffin joints. However, considerable variation
between subjects evident in the carpal and coffin joints traces (Figures 4.1.3 and
4.1.4) may have compromised the statistical power.

Figure 4.1.3 — Carpal joint traces for “mean sound” (solid black line) and the 5

trials for lame (solid gray lines; above) and contralateral (gray lines with circles;
below) forelimbs of the lame horses (group 1).

200 A . --..-- M... .-.-..-.--...----_--- --.-..-..._- .. ..

 

    

 

 

 

 

 

 

 

160 - ;.
12° 1 Stance _
1,? 100 . . . . .
g o 20 40 60 80 100
U)
8
E 200 \‘ g
8’ . .
< i 1 .- ‘. 7:;59'7': T-‘i 8?.” SWIn
. -- If)?
- g «1:? I
140 ~ “ , _. , flag 3
' 9 ~. -I.- .
"a” . .2
120 - v/
Stance 5...,
100 . . . . .5
0 20 40 60 \ 80 100
Time (% Stride)

Arrows indicate the events selected for analysis (maximum extension in stance
phase and maximum flexion in swing phase)

51

Figure 4.1.4 — Coffin joint traces for “mean sound” (solid black line) forelimb of
the control horses (group 2) and the 5 trials for lame (solid gray lines; above) and
contralateral (gray lines with circles; below) forelimbs of the lame horses (group

1).

 

260 w. _..., ._ “m.-. - ”-22.;

I
240 . wing
.---— ~_ ~
/ .‘
r, | h ,v
I._,"" xv . ‘.
3. ‘\ Q. ' ‘.
t . t
'1 .’ ’~_ ‘. __ ..,
C a. ' “- ‘ I I
I
s I . , a
I ‘ r .'
\ ~ - a. :

 

 

 

 

 

O 20 40 60 80 100

 

260
240 -
220 ~
200
180 '
160 .
140

120 Stance

100 l T I T T T
O 20 40 60 80 1 00

Angle (Degrees)

 

 

 

 

Time (% Stride)
Arrows indicate the events selected for analysis (maximum extension in stance
phase and maximum flexion in swing phase)
Single-subject analysis detected differences in carpal and coffin joints
between lame and contralateral forelimbs of subjects from group 1 (Table 4.1.1).

There was no standard movement pattern between lame and contralateral

52

forelimbs. Indeed, the relationship between the forelimbs was remarkably
individual even for the subjects with the same type of injury (e.g. subjects 1, 2

and 3; see Figure 4.1.5).

Figure 4.1.5 - Carpal maximum flexion (above) and coffin maximum extension
(below) angles : SD for lame (gray shading) and contralateral (striped shading)
forelimbs of the lame horses (group 1).

1304

120 . ---------------- -

  

7

  

\x‘ {-

 

1104

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

8‘
a) 100 -
h 1
a)
a)
9.
2 =I=
g) 240 -
<
230 ~
220 -. ff
210 .
200 -
190 '
1 2

4 5

3
Subjects

* Significant single-subject difference between lame and contralateral forelimbs

(p<0.05)
Dashed horizontal lines = “mean sound” forelimb

53

In regard to stride variables, both “mean Iame” and “mean contralateral”

forelimbs demonstrated a significantly longer (p<0.05) stance phase duration

compared to the “mean sound” forelimb. Along with longer stance phase, lame

horses also showed a significantly shorter (p<0.05) stride length than the “mean

sound” forelimb (Table 4.1.3). Three subjects (horses 2, 3 and 5) demonstrated

single-subject differences in stance duration. Stance duration was significantly

longer (p<0.05) on the lame forelimb compared to the contralateral forelimb in

subjects 2 and 5, while subject 3 showed the contrary.

Table 4.1.3 — Mean (SD) stride duration, stride length and stance duration for
lame (L) and contralateral (C) forelimbs of the lame horses (group 1), and for the
“mean sound” forelimb of the control horses (S; group 2).

 

 

 

 

Subject Stride duration (3) Length (m) Stance (% of stride)
L c L c L c
1 0.70(0.01)0.69(0.01) 2.18(0.05) 2.17(0.04) 47(002) 52(001)
2 0.68(0.01)0.66(0.01) 2.20(0.02) 2.17(010) 53(0.00)* 48(0.02)
3 0.61(0.01)0.63(0.01) 2.02(0.01) 2.01(0.03) 46(0.00)* 54(000)
4 0.70(0.01)0.71(0.00)2.19(0.04) 2.15(0.04) 51(000) 49(004)
5 0.72(o.00)0.72(0.00) 2.20(0.05) 2.15(0.04) 47(0.01)* 46(0.00)
wGroup1 088(0.05)0.89(0.04)2.18(0.07)a2.13(0.07)b 48(003)a 50(0.03)°
Group 2 (S) 0.71 (0.03) 2.28 (0.09) 43 (0.01)

 

 

 

4 Significant single-subject difference between L and C forelimbs (p<0.05)
3 Significant group mean difference between L and S forelimbs (p<0.05)
b Significant group mean difference between C and S forelimbs (p<0.05)

54

4.2. GROUND REACTION FORCES AND IMPULSES

Overall, the “mean lame” forelimb demonstrated significantly (p<0.05) smaller

peak vertical GRF than the “mean sound” forelimb (Table 4.2.1; Figure 4.2.1).

Table 4.2.1 — Mean (SD) peak vertical, braking and propulsive GRFs for lame
(L) and contralateral (C) forelimbs of the lame horses (group 1), and for the
“mean sound” forelimb of the control horses (S; group 2).

 

Subject

2
3
4
5

Peak Vertical Force
(N-kg“)

Peak Braking Force Peak Propulsive Force

(N-k9'1)

(N-k9'1)

 

L C

L C

 

 

 

 

7.36(O.1) 7.48(0.2)
9.34(0.1)*10.55(0.2)
9.36(0.1) 9.17(0.3)
8.09(0.1)*10.51(0.1)

-0.89(0’.2”) -1 .27(0.5)
-1.37(0.3)*-0.56(0.2)
~0.71(0.4)*-1.20(0.1)
_1.27(0.2)*-0.89(0.1)
-0.83(0.1)*-1.35(0.1)

0.72(0.1) 0.98(0.3)’~
033(01):. 0.84(0.1)
095(01): 0.54(0.1)
052(01):: 1.02(0.1)

071(00):: 0.92(0.0)

 

Group 1 788(22)a 8.85(2.1)

-1.01(0.3) -1.01(0.4)

0.64(0.1) 0.86(0.1)

 

Group 2 (S)

10.78(1.0)

 

-1.25(0.2)

 

 

 

 

0.92(02)

 

* Significant single-subject differences between L and C forelimbs (p<0.05)
3 Significant group mean differences between L and S forelimbs (p<0.05)

55

Peak braking and propulsive GRF peaks showed no group mean differences

within group 1 or between groups 1 and 2 (Table 4.2.1, Figure 4.2.1).

Figure 4.2.1 - Vertical (above) and longitudinal (below) ground reaction force
traces for “mean lame” (solid gray line) and “mean contralateral” (gray line with

circles) forelimbs of the lame horses (group 1), and for “mean sound” (solid black
line) forelimb of the control horses (group 2).

12-

 

 

 

O 20 4O 60 80 100

Force (N/kg)

 

 

 

Time (% Stance)

Arrows indicate the events selected for analysis (peak vertical, peak braking and

peak propulsive)
* Significant group difference between “mean lame” and “mean sound” forelimbs

(p<0.05)
56

Subjects 2 and 4 did not show significant single-subject differences in
vertical GRF peak between lame and contralateral forelimbs (Figure 4.2.2). A
high variability in the longitudinal GRF trace between subjects was observed,
with no standard longitudinal GRF pattern between lame and contralateral
forelimbs (Figure 4.2.2). Hence, single-subject analysis was more effective in
detecting significant differences (p<0.05) between lame and contralateral
forelimbs due to the elimination of inter-individual variability in the analytical
procedure. The lame forelimb of subjects 3 and 5 demonstrated a significantly
lower (p<0.05) braking GRF peak, while the lame forelimb of subjects 2 and 4
showed the contrary (greater braking; p<0.05). The longitudinal GRF showed
three distinct patterns. In subjects 2 and 4 the values were higher in the lame
limb, which resulted in more braking and leSS propulsion in the lame limb. In
subject 3, the values were lower in the lame limb, which resulted in less braking
and higher propulsive. ln subjects 1 and 5, the lame limb showed less braking

and less propulsion than the contralateral.

57

Figure 4.2.2 — Vertical and longitudinal ground reaction force traces for lame
(solid gray line) and contralateral (gray line with circles) forelimbs of the lame

horses (group 1).

 

 

 

 

10 ~ 0; - . ..
8 - A .0 . "N
6 q I. .‘o 5 ..°. 5 I
4 7 3". . 3‘ ’0'? 71.. ‘15:. .’
2 ‘I '1'"; -1.5 J o
0 ' 1 r I r \r '2 d
0 20 40 50 30 100 0 20 40 60 80 100

 

(3th
l I
E: o
Nmamomam
‘.
0"}.
1‘

A

I I J

/

 

ll_ll‘

(D
O
_.L
O
O
O
N
O
A
O
00
O
.5
O
O

 

 

 

 

 

 

 

 

 

 

 

6
¥ 12 1 a: 1.5 — ...
E 10 ~ 1 — ~ 3
(D 8 7 . ‘ o 5 J A
g 6 _ ..‘x .0 g _" 3': I. . I 1
u- 4 _ o'fl f... '1 " 1’; ‘0~ ’\/
2 2’ \ -1 5 — " ‘r' a:
0 1 I r 17 r -2 ‘
O 20 40 60 80 100 0 20 4O 60 80 100
12 7 1.5 1 *
10 1 1 ‘ , 4
8 T \ 03 q r r l/T r
2 : .0. $3. ”0.5 J...~~ll:~ .Q/
,J -1 ‘ *
2}/ \\\ 451
O I r T I T -2 T
0 20 40 60 80 100 0 20 40 60 80 100
12 _ * 1 5
1° ‘ '1 .7. 5
S i r"... ‘0'? i i M/
'5 - 05‘. ' . .
2 . x."- -1.5 ' ' '- 4
0 . I r 1 l ‘\1 '2
0 20 40 60 80 100 0 20 40 60 80 100
Time (% Stride)

* Significant single-subject differences between lame and contralateral

forelimbs

58

In the group analysis of impulses, the “mean Iame” forelimb had
significantly lower (p<0.05) vertical impulse than the “mean sound” forelimb. No
significant group differences were observed in the braking and propulsive
impulses (Table 4.2.2).

Table 4.2.2. Mean (SD) vertical, braking and propulsive impulse of lame (L) and

contralateral (C) forelimbs of the lame horses (group 1), and for “mean sound” of
the control horses (8; group 2).

 

 

 

Vertical Impulse Braking Impulse Propulsive Impulse
Subject I
(Ns.kg") (Ns.kg‘1) I (Ns.kg")
L c I L c L c
” .___ 1 1.68(0.0)* 2.14(0.0) -0.13(0.0) -0.17(0.1) 0.08(0.0) 0.09(0.0)
2 1.75(0.0) 1.67(0.1) -0.24(0.0)* -0.09(0.0) 002(00):: 0.09(0.0)

3 1.58(0.0)* 1.83(0.1) -0.08(0.0)* -0.15(0.0) 0.09(0.0)ar 0.04(0.0)
4 1.85(0.0)* 1.91 (0.0) -0.16(0.0)*-0.08(0.0) 0.04(0.0): 0.10(0.0)

5 1.68(0.0)* 1.94(0.0) -0.12(0.0) -0.11(0.0)0.07(0.0)* 0.09(0.0)

 

”Growl” 1.70(0.1)a 1139(02) -0.15(0.1)-0-12(0.1) 0.06(0.0) 0.08(0.0)

 

 

Group2(S) 1.92(0.1) l -0.12(0.0) I 0.08(0.0)

 

* Significant single—subject differences between L and C forelimbs (p<0.05)
3 Significant group mean differences between L and S forelimbs (p<0.05)

Single subject analysis detected significant differences (p<0.05) in the
vertical, braking and propulsive impulses between lame and contralateral
forelimbs in the majority of the subjects. Subjects 1, 3, 4 and 5 showed
significantly smaller (p<0.05) vertical impulse on the lame limb compared to the

contralateral. Similarly to braking and propulsive GRF peaks, braking and

59

propulsive impulses showed individual patterns that differed between horses
(Figure 4.2.3).

Figure 4.2.3 - Braking impulse i SD for lame (gray shading) and contralateral
(striped shading) forelimbs of the lame horses (group 1).

 

55
0
U1

 

 

i”. .b
01 _;

 

Force (N/kg)

.5
N

 

 

 

 

 

Subjects

* Significant single-subject difference between lame and contralateral forelimbs
(p<0.05)
Dashed horizontal lines = “mean sound” forelimb

The relationship between magnitude of propulsive and braking impulses

on subjects 2, 3 and 4 was similar to that of propulsive and braking GRF peaks.

4.3. BIOMECHANICAL VARIABILITY

Variability for “mean lame”, “mean contralateral” and “mean sound”
forelimbs was calculated by averaging the standard deviations obtained from the
5 trials of each subject. A “mean SD" was then compared between conditions

(lame, contralateral and sound). Variability differences between conditions were

60

observed for some of the kinematic and GRF variables measured. Overall, the
sound condition demonstrated significantly greater (p<0.05) biomechanical
variability than the lame condition. For instance, variability of the maximum
fetlock joint extension for the lame horses (“mean SD lame” and “mean SD
contralateral”) forelimbs was significantly smaller (p<0.05) than for the sound
horses (“mean SD sound”). Lame horses also showed significantly smaller
(p<0.05) variability in stride duration than the sound horses.

With regard to GRF variables, the variability in the propulsive GRF peak
for the lame horses (“mean SD lame” and “mean SD contralateral” forelimbs)
was significantly smaller (p<0.05) than for the sound horses (“mean SD sound”

forelimb).

61

4.4. GRAPHICAL REPRESENTATIONS OF BIOMECHANICAL DATA

4.4.1. Kinematics

Differences in joint angle patterns were illustrated initially through the use
of angle-time diagrams. Figure 4.1.1 (p. 48), which compares fetlock angles for
the “mean sound”, “mean lame” and “mean contralateral” conditions, shows that
there are differences throughout the stride, especially in the stance phase, when
the maximum extension angle of the “mean lame” forelimb was significantly
smaller (p<0.05) than the “mean sound”. Analysis of kinegrams from individual
subjects provides an impression of the movement pattern of the entire limb and
the coordination between the joints. In addition to highlighting differences in the
movement patterns, kinegrams show overall differences in maximum height and
length displacement of the forelimbs during the stance phase, that were
quantified by including a numerical scale on the y and x axes. For instance,
subject 1 held the lame forelimb higher (greater maximum height) and had a
greater length displacement in the lame forelimb during stance phase than in the
contralateral forelimb (Figure 4.4.1.1). Although some horses showed similar
length displacement between the forelimbs (e.g. subject 3), the lame forelimb
could still be identified due to the differences in maximum height of the elbow
marker between lame and contralateral forelimbs, where the lame forelimb was
held higher than the contralateral, especially at the beginning of stance phase

(Figure 4.4.1.1).

62

Figure 4.4.1.1 — Kinegram of lame (dashed gray line) and contralateral (solid
black line) forelimbs of subject 1 (above) and subject 3(below).

Height (m)

0.8 r

0.6 F

0.4

0.2 ~

0.8

0.6 r

0.4 r

0.2

 

 

 

 

 

 

 

Lame limb held higher than
contralateral
Lame limb with
greater length
. displacement than
contralateral
~0i.6 -0 4 -012 0 0.12 0: 02 08
Lame limb held higher than
contralateral
0&6 -0.4 -01.2 0 0.12 0.14 0f6 078
Length (m)

63

 

4.4.2. Ground reaction forces

Vertical force-time diagrams allowed easy recognition of the lame forelimb
in subjects 1, 3 and 5, but not in subjects 2 and 4 (see section 4.2; Figure 4.2.2).
Differences in braking and propulsive longitudinal GRFs were observed between
lame and contralateral limbs, but the differences were not consistent between
horses, making it difficult to determine which was the lame forelimb (see section
4.2; Figure 4.2.2).

Force vector diagrams allowed a more detailed visualization of differences
between lame and contralateral forelimbs in vertical, braking and propulsive
GRFs simultaneously (Figure 4.4.2.1). Inclusion of the SD for peak vertical,
braking and propulsive forces for sound horses, which are shown as dashed
lines on the force vector diagrams, facilitated the detection of lameness. The
smaller peak vertical force in the lame forelimb compared to the contralateral
one, which had been observed with the force-time diagram, was clearly identified
with the force vector diagrams. Even when there was no difference in peak
vertical force between lame and contralateral forelimbs (subjects 2 and 4), the
longitudinal GRF, especially its braking component, showed a distinct difference
between lame and contralateral forelimbs. In these horses, the force vectors
leaned to the left indicating a larger braking force in the first-half of the stance
phase. Moreover, the propulsion vectors for the lame forelimb were visibly lower
than the contralateral forelimb, and did not reach the expected sound forelimb
range (SD for peak propulsive force of group 1) represented by the vertical

dashed lines on the right side of the force vectors. The time of mid stance of the

64

lame limb, shown by the dashed line in between the solid force vector lines, also
leaned noticeably towards the left (braking direction) compared to the

contralateral forelimb.

65

Figure 4.4.2.1 — Force vector diagrams of GRFs from lame and contralateral
forelimbs of the lame horses (group 1).

1

5

12 —. 'L—armejmrg ‘7

 

Contralateral limb
1‘ i T I" " T f T '

 

{I‘i
HI

_;
O
. 1

 

Y

 

L
r
I
I
I
I
I
I
I
I
r
""r
I
I
I

a)

 

Vertical force (ng'1)
m
‘*-\

h.

 

I
I
l
l
l
l
l
l
l

 

I I \ ll 1
-2 o' 2 -2

 

l
........
i I

Vertical force (N.kg‘ 1)

 

 

-2 “MT 707 WNW“?
Horizontal force (N.kg'1)
12i I_La|m_ell_mt~|3_#

 

Vertical force (N.kg'1)

 

Horizontal force (N.kg‘1)

 

 

 

 

Contralateral limb

 

L. Lame limb. @ntralaterel ILmb

 

 

I WW I 1

‘ i a I s I
10' i I « I I

-L--.'- L l L I

 

 

 

 

I
t i i 1
|

 

 

A;

Vertical force (N.kg'1)

 

 

 

 

 

 

 

2 -2 o 2
4 Horizontal force (N.kg'1)

 

12-.. amen?

I
I 1
-4.--, ...... r}---¢
I I

Enmgatlimb

I
'1'" ------- I-I-n-
l I I

----------

 

>,L.‘, . ’ 1‘

.4- ' .m. z
o - _ ..a - ‘ fl...-
.\r f .
.. - 7‘ ., .
I
~ - .x 1...: - q

  

 

Vertical force (N.kg‘1)
a:

 

 

 

 

 

 

 

 

-2 0 2 -2 7
Horizontal force (ng'1)

Sequential GRF vectors (every 4% of stance
phase) form a fan-shaped profile. 50% time is
indicated by a dashed line between vectors;
SD from sound forelimbs is shown at the top
of the graph for the vertical GRF, at the left of
the graph for the braking GRF and at the right
of the graph for propulsive GRF.

66

CHAPTER 5

DISCUSSION

Muskuloskeletal injuries are the most common condition afflicting
racehorses (Rossdale et al., 1985) and the most common reason that horses in
training failed to race (Jeffcott et al., 1982; Rossdale et al., 1985). Horses often
manifest musculoskeletal injury by changing the movement and loading profile of
the limbs, a condition called lameness (Morris et al., 1987). Assessment of
locomotor pattern is usually based on subjective measurements gathered by the
human eye with the images being processed by the human brain (Schamhardt et
al., 1993). Although the eyes of an experienced clinician are the most convenient
diagnostic method for lameness (Perry, 1992), this subjective assessment can
produce misleading judgments (Leach, 1987; Ratzlaff, 1989). Two
complementary biomechanical approaches can objectively measure gait in
humans and animals: kinematics and kinetics. Kinematics provides information
about the geometry of motion, such as joint angular patterns, and kinetics
describes the external and internal forces, such as ground reaction forces and
joint torques respectively (Leach & Crawford, 1983; Galisteo et al., 1997; Barrey,
1999; Nigg, 1999).

Joint angular patterns are an important indicator of physiologic and
locomotor capacity (Back et al., 1994; Holmstrom et al., 1994). Joint angular

patterns change with induced lameness (Buchner et al., 1996; Keegan et al.,

67

1997), where reduction in the maximum extension of the fetlock joint has been
described as a consistent finding in lameness models, such as induced carpal
lameness (Back et al., 1993), sole pressure-induced lameness (Buchner et al.,
1996) and lameness by induced superficial digital flexor tendinitis (Clayton et al.,
2000). The fetlock joint extends during weight acceptance then flexes as the limb
pushes off the ground. The lame horses in this study showed a reduction in
maximum extension of the fetlock joint, with the lame forelimb (“mean lame”)
being 19° less extended than the sound forelimb (“mean sound”). This is typical
of a supporting limb lameness, in which the horse reduces the load on the painful
limb, with a corresponding reduction in the ground reaction forces (Buchner et
al., 1996; Buchner, 2001). The fetlock joint angle-time curve during the stance
phase has been reported to resemble the vertical ground reaction force pattern in
both lame (Buchner et al., 1996) and sound conditions (Riemersma et al., 1988;
Ratzlaff et al., 1990). This relationship between vertical GRF and fetlock
extension was apparent in this study in which a decrease in fetlock maximum
extension of the “mean lame" forelimb was associated with a significant reduction
in vertical GRF peak (3 N/kg lower vertical GRF peak than “mean sound”;
p<0.05). As suggested by Back et al. (1993), the vertical GRF represents a
supporting limb component of lameness, which was also evident in this study.

A reduction in maximum flexion of the coffin joint has also been
associated with induced supporting limb lameness (Buchner et al., 1996).
However, no significant group mean differences in maximum coffin flexion were

observed in this study. This lack of change in coffin flexion could be due to the

68

high variability in coffin joint angular patterns between horses, which may have
been a consequence of the different injury types and severities. In addition,
previous studies describe the coffin joint as having one of the highest inter-
individual variabilities (Deguerce et al., 1997). Moreover, age (Galisteo et al.,
2001) and conformation (Deguerce et al., 1997; Butcher et al., 2002) have also
been shown to increase inter-individual variability of kinematic measures. For
instance, 2 year-old Thoroughbreds have been shown to have a quicker rate of
fetlock joint dorsi-flexion compared to older Thoroughbreds (3, 4 and 5 year-old)
(Butcher et al., 2002). Even though no studies have been performed regarding
an older population of horses, this study showed no apparent reduction in fetlock
range of motion with increasing age in the control group (age = 12 i 5.4), thus
age was not considered a confounding variable in this study. Although
differences in conformation between the subjects and the control group (2
Warmbloods and 3 Thoroughbreds) could have affected the analysis of kinematic
variables, the data were found to be normally distributed with homogeneous
variance, thus discounting this as a confounding effect and not disputing the
statistical findings.

Findings regarding a decrease in carpal flexion have been controversial.
Maximum flexion of the carpus has been described to decrease with induced
carpal lameness and with naturally occurring navicular disease (Back et al.,
1993; Keegan et al., 1997), but no changes were detected with sole pressure-

induced forelimb lameness (Buchner et al., 1996). As found by Buchner et al.

69

(1996), maximum carpal flexion remained unchanged with lameness in this
study.

Single-subject analysis of joint angular patterns was more powerful than
group analysis in detecting intra-individual differences between lame and
contralateral forelimbs due to the abolishment of the effects of individual variation
on the statistical analysis. The amount of reduction in fetlock extension in the
lame limb varied between subjects and did not necessarily resemble the pattern
of the vertical GRF contradicting Buchner et al. (1996). For instance, subject 4
demonstrated a remarkable reduction in fetlock extension (26°) on the lame limb,
but the vertical GRF peaks for both lame and contralateral forelimbs were within
normal limits (chapter 4, Figure 4.4.2.1.). Single-subject analysis was also able to
detect angular changes in maximum extension of the coffin joint between lame
and contralateral forelimbs, in particular the unexpected increase in maximum
extension of the coffin joint manifested by subjects 1, 2, 4 and 5. Single-subject
differences in coffin flexion were not found consistently, but there was an overall
trend for maximum coffin flexion to increase in the lame limb, which was also an
unexpected finding. This overall increase in angular motion of the coffin joint
might be specific to fetlock joint injury; however, further studies need to be
developed to investigate this hypothesis. In regard to the carpal joint, single-
subject differences in maximum extension and flexion between lame and
contralateral forelimbs were also diverse, with no consistent change shown by

the lame limb.

70

The longer stance phase duration shown by the lame horses in both lame
and contralateral forelimbs in comparison to the sound horses was previously
reported in horses with sole pressure-induced lameness (Buchner et al., 1996;
Galisteo et al., 1997; Keegan et al., 2000), and in a single horse with chronic
sesamoiditis (van Weeren et al., 1993). The increase in stance duration has
been suggested as a means of reducing the peak vertical force by distributing the
force over a longer duration (Galisteo et al., 1997; Buchner et al., 1996; Keegan
et al., 2000). Stance duration is dependent on speed, and if not controlled, speed
of the horse may voluntarily decrease with lameness (Riemersma et al., 1988). In
this study, trotting speed was controlled (3.0 to 3.3 m.s'1) and an increase in
stance duration was considered to be an indicator of supporting lameness of
moderate degree. Besides the longer stance phase found in the lame horses,
stride length was significantly reduced compared to the sound horses (2.13 m
against 2.28 m; p<0.05). which supported the findings from induced supporting
limb lameness models (Back at al., 1993; Deuel et al., 1995; Buchner et al.,
1995; Galisteo et al., 1997).

In regard to the ground reaction forces, the reduction in the vertical GRF
peak in the lame forelimb has been associated with a decrease in the duration of
the aerial phase of the stride (Morris & Seeherman, 1987). This reduced aerial
phase in a smaller vertical displacement of the center of mass cause a decrease
in vertical GRF peak. Analysis of longitudinal GRFs provided complementary
information for the identification of the lame limb. A reduction in the braking GRF

peak has also been related to lameness and attributed to the inability of the

71

horse to decelerate the lame limb (Morris & Seeherman, 1987). In contradiction
to previous findings, group analysis did not demonstrate a decrease in braking
GRF peak in the lame condition compared to the sound condition. Instead, the
pattern of braking and propulsive GRF peaks showed no standard pattern
between subjects.

The lack of single-subject differences in vertical GRF peak between lame
and contralateral forelimbs on subjects 2 and 4 was unexpected and might be
indicative of a lesser severity of lameness, bilateral lameness, or even a different
mechanism of redistribution of ground reaction forces between the limbs. Hence,
the use of vertical GRF peak as a supporting limb component of lameness, as
suggested by Back et al. (1993), may be misleading if used in isolation to
distinguish between lame and contralateral forelimbs, as described in previous
studies (Williams et al., 1999). Single-subject analysis was more powerful than
the group analysis in detecting differences in longitudinal GRF and impulse
between lame and contralateral forelimbs due to the elimination of the effects of
individual variation in the statistical analysis. The patterns of longitudinal GRF
between lame and contralateral limbs were diverse. The significant decrease
(p<0.05) in braking GRF peak observed in the lame forelimb of subjects 3 and 5
was an expected finding and has been related to the inability of the limb to
decelerate (Morris & Seeherman, 1987). In humans, a protective posturing to
protect joint structures has been described as a physiological reaction to pain. As
a result of this protective posturing, muscle atrophy may occur (Perry, 1992). In

horses with lameness, this protective posturing may also take place causing

72

muscle atrophy and consequent muscle weakness leading to the inability of the
horse to decelerate the lame limb.

Even though subjects 3 and 5 showed similar pattern of braking force in
their lame limb, the propulsive component of their longitudinal force was different,
showing an increase in the lame limb of subject 3, and a decrease in the lame
limb of subject 5. The increased propulsive force (p<0.05) in the lame limb of
subject 3 was unexpected and may be due to the horse’s effort to maintain a
constant body velocity. Supporting the individual variation in longitudinal GRF,
the braking force of subjects 2 and 4 was unexpectedly increased in the lame
limb, contrary to the pattern observed in subjects 3 and 5. This greater braking
force may be due to either a variation in the severity of lameness, or perhaps a
different compensatory mechanism of adaptation to lameness. Indeed, in
addition to the unusual increase in braking force, the lame forelimb of subjects 2
and 4 also showed a remarkable decrease in propulsive GRF peak. Therefore,
the lame limb was decelerating effectively but it was unable to push the lame
limb off the ground effectively. This phenomenon may be associated with the
type of injury to the fetlock joint. The fetlock joint is known to be the main site for
elastic energy storage and release during the stance phase (Clayton et al.,
1998). As the limb is loaded in early stance, the fetlock joint extends, and
stretches the suspensory ligament, superficial and deep digital flexor tendons,
and their respective accessory ligaments. Elastic energy is stored during loading
and is later released as the limb is unloaded, thereby conserving energy. Horses

with supporting limb lameness may have the loading phase of the limb

73

compromised; therefore, the storage of elastic energy is affected. Although
subjects 2 and 4 did not have significant differences in vertical GRF peak
between lame and contralateral forelimbs, the vertical force was abnormally low
in both limbs. The resulting decrease in elastic energy release may have caused
a reduction in force available for push off, leading to a reduced propulsive GRF
peak at the end of the stance phase.

Interestingly, the variability of some kinematic (maximum fetlock extension
and stride duration) and GRF (propulsive GRF peak) variables was much lower
in the lame condition compared to the sound condition. Peham et al. (2001)
suggested that horses with orthopedic pain keep stride variability low to optimize
compensatory mechanisms to reduce pain in the affected limb. Variability is said
to offer flexibility to adapt to perturbations (Holt et al., 1995) and is essential to
the changing coordination patterns during locomotion (van Emmerick et al.,
1999). In humans, van Emmerik et al. (1999) demonstrated that individuals with
patellofemoral pain displayed reduced coordination variability among joint
couplings of the painful limb, with the individuals avoiding painful coordination
patterns (Heiderscheit et al., 2002). Force variability has also been suggested to
increase as the level of force produced increases (Jenkins, 1947; Noble &
Bahrick, 1956; Provins, 1957), so the decrease in variability of the propulsive
GRF peak could be explained by the overall reduction of GRF manifested by the
lame horses. The reduced variability in some of the kinematic and GRF
measures supports previous studies and suggests the use of movement

variability as a discriminating measure and possible clinical tool.

74

 

Illustration of kinematic measures, such as joint angular patterns, further
detection of the lame limb was achieved with the use of angle-time diagrams.
The stance phase is subjected to coordination changes with supporting limb
lameness (Stashak, 2002), thus the use of kinegrams from the stance phase
represented the motion of the entire limb and highlighted asymmetries and
changes in coordination between the limbs. As suggested by Back et al. (1995),
the association of joint angle-time diagrams with the movement of the entire limb
may facilitate the interpretation of kinematics. This study supports the fact that
graphical displays of kinematic data enhance understanding and suggests that
the incorporation of kinegrams from the stance phase facilitates the identification
of lameness. In regard to ground reaction forces, this study suggests that in
addition to analysis of the vertical GRF pattern, the longitudinal GRF should also
be used for identification of gait abnormalities. The lack of significant differences
in the peak vertical GRF between lame and contralateral forelimbs in subjects 2
and 4 supported the findings from previous studies showing that asymmetry in
peak vertical force may be misleading if used in isolation to distinguish between
normal and abnormal gait patterns (Merkens & Schamhardt, 1988; Williams et
al., 1999). Individual responses in ground reaction forces, especially in the
braking component of the longitudinal GRF, made it difficult to show statistical
significance between limbs; hence in addition to the force-time diagrams, a
simpler method to measure lameness is required.

The use of force vector diagrams to compare lame and contralateral limbs

proved useful identifying the lame limb. Vector diagrams have been used in

75

 

human gait analysis to illustrate pathological gait patterns (Rozendal et al., 1985;
Khodadadeh, 1988; Baumann et al., 1992), but their use in horses as a potential
diagnostic tool for lameness evaluation has been limited (Merkens et al., 1985;
Merkens et al., 1988). Force vector diagrams are beneficial as they enable more
than one variable (i.e. peak vertical, braking and propulsive GRFS) of both limbs
to be observed and analyzed simultaneously. The inclusion of lines illustrating
the SD for the peak forces based on data from sound horses was also helpful for
detecting abnormal forces in the contralateral limb. Changes in the movement
pattern in response to pain have been described to redistribute loads through
other joints in the body and boost the development of secondary injuries (Whiting
& Zemicke, 1998). Detection of abnormal forces in the contralateral limb is
imperative because it could contribute to the development of secondary injuries.
Similar to what was demonstrated in previous studies (Schamhardt et al., 1993;
Merkens et al., 1993), force vector diagrams facilitated evaluation of left versus
right symmetry, or asymmetry of specific points on the force time curve. Indeed,
these diagrams were easier to understand than more complex approaches, such
as the use of principal component analysis (Williams et al., 1999). Left versus
right symmetry of GRFs in equine locomotion is relevant to athletic ability (Dow et
al., 1992), and identification of force vector asymmetries consistent with
pathological gait characteristics might assist practitioners to establish a diagnosis

and prognosis, and to assess the therapeutic response.

76

 

CHAPTER 6

CONCLUSIONS

Refinement of the diagnosis of pathological gaits, study of the etiology of

muskuloskeletal disorders and evaluation of treatment may be improved through

objective quantitative analyses (Keg et al., 1994). This study has demonstrated

that:

Kinematic measurements can detect gait abnormalities in horses, with
fetlock joint maximum extension, stance phase duration and stride length
being the main indicators of fetlock joint lameness.

Vertical ground reaction forces tend to show an overall reduction in the
lame limb.

Longitudinal GRFs are highly variable between horses even when the
etiology of lameness is similar.

Group statistical analysis was not an effective method of characterizing
lameness in horses with fetlock joint injuries due to inter-individual
vafiafion.

Single-subject analysis offers a powerful intra-individual analytical method
to determine asymmetries between lame and contralateral limbs and
therefore, detecting lameness.

Lame horses reduce their intrinsic movement variability to avoid painful

coordination patterns.

77

Movement variability may be used as a discriminating measure and
possible clinical tool. Based on this information, clinicians can incorporate
measures of movement variability to aid in the design of appropriate
treatment programs directed at these deficits.

Graphical displays such as angle-time diagrams, kinegrams, force-time
diagrams and force vector diagrams facilitate the interpretation of
biomechanical data and can be used as simple tools to quantify lameness

objectively.

78

 

REFERENCES

Back W, Bameveld A, Weeren PR van, et al. (1993) Kinematic gait
analysis in equine carpal lameness. Acta Anat, 146(2-3):86-89.

Back W, Bameveld A, Schamhardt HC, Bruin G, Hartman W (1994)
Longitudinal development of the kinematics of 4-, 10-, 18- and 26- month-old
Dutch Warmblood horses. Equine Vet J, (Suppl. 17):3-6.

Back W, Schamhardt HC, Hartmann W, et al. (1995) Kinematic
differences between the distal portions of the forelimbs and hind limbs of horses
at the trot. Am J vet Res, 56(11):1522-1528.

Back W, Schamhardt HC, Savelberg HHCM, et al. (1995) How the horse
moves 1. Significance of graphical representations of equine forelimb kinematics.
Equine Vet J, 27(1):31-38.

Back W, Schamhardt HC, Bameveld A (1996) The influence of
conformation on fore and hind limb kinematics of the trotting Dutch Warmblood
horse. Pferdeheilkunde, 12(4):647—650.

Barlow HD, Herson M (1984) Single case experimental design: Strategies
for studying behavioral change. New York: Pergamon Press, 1-35.

Barrey E (1999) Methods, applications and limitations of gait analysis in
horses. Vet J, 157(1):7-22.

Bartel DL, Schryver HF, Lowe JE, Parker RA (1978) Locomotion in the
horse: a procedure for computing the internal forces in the digit. Am J vet Res,
39:121-1727.

Bartlett RM (1997) An introduction to Sports Biomechanics. London: E 8
FN Spon.

Bates BT, James SL, Osteming LR, Mason BR (1979) Functional
variability within the support phase of running. Medicine and Science in Sports
and Exercise, 11:328-331.

Bates ET (1989) The influence of running velocity and midsole hardness .
on external impact forces in heel-toe running. J Biomech, 22:963-965.

79

 

Bates BT (1996) Single-subject methodology: an alternative approach.
Med Sci Sports Exerc, 28:631-638.

Baumann JU, Schar A, Meier G (1992) Forces and turning moments of hip
and knee joints in locomotion. Orthopade, 21 :29-34.

Bernstein N (1967) Coordination and regulation of movement. Oxford:
Pergamon Press.

Bjorck G (1958) Studies on the draught forces of horses: development of a
method using strain gauges for measuring between hoof and ground. Acta Agric
Scand (Suppl 4).

Bogert AJ van den, Schamhardt HC (1993) Multi-body modeling and
simulation of animal locomotion. Acta Anat, 146295-102.

Buchner HHF (2001) Gait adaptation in lameness. In: Back W, Clayton
HM (eds). Equine locomotion. London: WB Saunders, 251-279.

Buchner HHF, Savelberg HHCM, Schamhardt HC, Bameveld A (1996)
Limb movement adaptations in horses with experimentally induced fore- or
hindlimb lameness. Equine Vet J, 28(1):63—67.

Buchner HHF, Savelberg HHCM, Schamhardt HC, et al. (1994)
Kinematics of treadmill versus overground locomotion in horses. Vet Quart
16:(Suppl. 2):887-890.

Buchner HHF, Savelberg HHCM, Schamhardt HC, et al. (1995) Bilateral
lameness in horses - a kinematic study. Vet Quart, 17(3):103-105.

Bukowiecki CF, Bramlage LR, Gabel AA (1987) In vitro strength of the
suspensory apparatus in training and resting horses. Vet Surg, 16(2):126-30.

Butcher MT, Ashley-Ross MA (2002) Fetlock joint kinematics differ with
age in thoroughbred racehorses. J Biomech, 35:563-571.

Cheney JA, Shen CK, Wheat JD, et al. (1973) Relationship of racetrack
surface to lameness in Thoroughbred racehorses. Am J Vet Res, 34:1285.

Clanton C, Kobluck CN, Robinson RA, et al. (1991) Monitoring surface
conditions of a thoroughbred racetrack. J Am Vet Med Assoc, 1982613.

80

Clayton HM (1986) Cinematography analysis of the gait of lame horses. J
Equine Vet Sci, 6:70—78.

Clayton HM (1987) Cinematographic analysis of the gait of lame horses 3
- fracture of the 3rd carpal bone. J Equine Vet Sci, 7(3):130—135.

Clayton HM (1988) Cinematographic analysis of the gait of lame horses 5
- fibrotic myopathy. J Equine Vet Sci, 8(4):297-301.

Clayton HM (1996) Instrumentation and techniques in locomotion and
lameness. Vet Clin N Am-Equine, 12(2):337.

Clayton HM, Almeida PE, Prades M (2002) Double Blind Study of the
Effects of Corta-le in Horses with Tarsal Degenerative Joint Disease. In:
Proceedings American Association of Equine Practitioners, 314-317.

Clayton HM, Lanovaz JL, Schamhardt HC, et al. (1998) Net joint moments
and powers in the equine forelimb during the stance phase of the trot. Equine Vet
J, 30(5):384-389.

Clayton HM, Schamhardt C (2001). Measurements and technique in
equine gait analysis. In: Back W, Clayton HM (eds). ln: Equine Locomotion,
London:WB Saunders. 55-76.

Clayton HM, Schamhardt HC, Willemen MA, et al. (2000) Kinematics and
ground reaction forces in horses with superficial digital flexor tendonitis. Am J Vet
Res, 61(2):191-196.

Clayton HM, Theoret CL, Barber SM, et al. (1998) Effects of carpal
synovectomy on stride kinematics of trotting horses. Vet Comp Orthopaed,
11(2):80-84.

Cohen ND, Peloso JG, Mundy GD, Fisher M, et al. (1997) Racing-related
factors and results of prerace physical inspection and their association with
musculoskeletal injuries incurred in thoroughbreds during races. J Am Vet Med
Assoc, 21 1(4):454-63.

Degueurce C, Pourcelot P, Audigié F, et al. (1997) Variability of the joint
patterns of sound horses at trot. Equine Vet J, (Suppl. 23):89-92.

81

Dow SM, Wilson AM, Goodship AE (1992) The symmetry of ground
reaction forces following racing as a predictor of wastage in Thoroughbred
racehorses, In: Proceedings. Annu Meet Eur Soc Biomech, 8.

Drevemo S, Dalin G, Fredericson l, Hjerten G (1980) Equine locomotion;
1. The analysis of linear and temporal stride characteristics of trotting
Standardbreds. Equine Vet J, 12(2):60-65.

Dufek JS, Barry TB, Stergiou N, James CR (1995) Interactive effects
between group and single-subject response patterns. Human Mov Sci, 14:301-
323.

Estberg L, Stover SM, Gardner IA, et al. (1996) Fatal musculoskeletal
injuries incurred during racing and training in Thoroughbreds. J Am Vet Med
Assoc, 208(1):92-96.

Fuller CJ, Simpson N, Mason N, Colbome GR (2002) Correlation of
lameness scores with ground reaction force symmetry, In: Proceedings Annu
Meet British Equine Vet Assoc.

Galisteo AM, Cano MR, Morales JL, et al. (1997) Kinematics in horses at
the trot before and after an induced forelimb supporting lameness. Equine Vet J,
(Suppl. 23):97—101.

Galisteo AM, Morales JL, Cano MR, et al. (2001) Inter-breed differences in
equine forelimb kinematics at the walk. J Vet Med A Physiol Pathol Clin Med,
48(5):277-285.

Geary JEJr (1975) The dynamics of the equine foreleg. Masters of
Mechanical and Aerospace Engineering, University of Delaware.

Gingerich DA, Auer JA and Fackelman GE (1979) Force plate studies on
the effect of exogeneous hyaluronic acid on joint function in equine arthritis. J Vet
Pharmacol Therap, 22291-298.

Goodship AE, Brow PN, Mac Fie HJH et al. (1983) A quantitative force
plate assessment of equine locomotor performance. In: Snow DH, Persson SGB,
Rose RJ (eds) Equine Exercise Physiology. Cambridge: Granta, 263-270.

Gray J (1968) Vertebrates. In: Animal Locomotion. London: Weidenfeld
and Nicolson, 3-18.

82

Hay JG, Reid JG (1982) The anatomical and mechanical bases of human
movement. Englewood Cliffs, NJ: Prentice-Hall.

Heiderscheit BC, Hamill J, van Emmerik REA ( 1999) Q angle influences
on the variability of lower extremity coordination. Medicine and Science in Sports
and Exercise, 31 :1313-1319.

Higgins JR (1977) Human movement: An integrated approach. St. Louis:
CV Mosby.

Hill AE, Carpenter TE, Gardner IA, et al. (2003) Evaluation of a stochastic
Markov-chain model for the development of forelimb injuries in Thoroughbred
racehorses. Am J vet Res, 64(3):328-337.

Hill TD, Carmichael D, Maylin G, et al. (1986) Track condition and racing
injuries in Thoroughbred horses. Come/l Vet, 76:361.

Homstrom M (1990) Variation in conformation of Swedish Warmblood
horses and conformational characteristcs of elite sport horses. Equine Vet J,
22(3):186—193.

Holmstrom M, Fredricson J, Drevemo S (1994) Biokinematic analysis of
the Swedish Warmblood riding horse at trot. Equine Vet J, 26(3):235-240.

Holt KG, Jeng S, Tacliffe R, Hamill J (1995) Energetic cost and stability
during human walking at the preferred stride frequency. J Motor Behavior,
27:164-178.

Hubert JD, Latimer FG, Moore RM (2001) Proximal sesamoid bone
fractures in horses. Comp Cont Educ Pract, 23(7):678-687.

Jeffcot LB, Rossdale PD, Freestone J, et al. (1982) An assessment of
wastage in Thoroughbred racing from conception to 4 years of age. Equine Vet J,
14:185.

Jenkins WC (1947) The discrimination and reproduction of motor
adjustments with various types of aircraft controls. Am J Psychology, 60:397-406.

Johnston C, Roepstorff L, Drevemo S, et al. (1996) Kinematics of the
distal hindlimb during stance phase in the fast trotting Standardbred. Equine vet J
28(4):263—268.

83

 

Kane AJ, Stover SM, Gardner IA, et al. (1996) Horseshoe characteristics
as possible risk factors for fatal musculoskeletal injury of Thoroughbred
rachorses. Am J Vet Res, 57:1147-1152.

Keegan KG, Wilson DA, Smith BK, et al. (2000) Changes in kinematic
variables observed during pressure-induced forelimb lameness in adult horses
trotting on a treadmill. Am J Vet Res, 61(6):612-619.

Keegan KG, Wilson DA, Wilson DJ, et al. (1998) Evaluation of mild
lameness in horses trotting on a treadmill by clinicians and interns or residents
and correlation of their assessment with kinematic gait analysis. Am J Vet Res,
11:1370-1377.

Keegan KG, Wilson DJ, Wilson DA, et al. (1997) Effects of anesthesia of
the palmar digital nerves on kinematic gait analysis in horses with and without
navicular disease. Am J Vet Res, 58(3):218-223.

Keg PR, Bameveld A, Schamhardt HC, et al. (1994) Clinical and force
plate evaluation of the effect of a high plantar nerve block in lameness caused by
induced mid-metatarsal tendonitis. Vet Quart, 162870-75.

Kelso JAS (1995) Dynamic Patterns: The Self-Organization of Brain and
Behaviour. Cambridge, MA: The MIT Press.

Khodadadeh S (1988) Vector (butterfly) diagrams for osteoarthritic gait: a
preliminary report. J Med Eng Techno], 12:15-19.

Khumsap S, Clayton HM, Lanovaz JL, Bouchey M (2002) Effect of walking
velocity on forelimb kinematics and kinetics. Equine Vet J, (Suppl. 34):325-9.

Kobluck CN (2003) Epidemioology of racehorse injuries. In: Diagnosis and

Management of Lameness in the Horse. Ross MW, Dyson SJ (eds). St Louis:
Saunders, 861-867.

Krook L, Maylin GA (1988) Fractures in Thoroughbred racehorses. Come/l
Vet, (Suppl. 11):11-33.

Leach DH (1987) Locomotion analysis technology for evaluation of
lameness in horses. Equine Vet J, 19(2):97-99.

Leach DH, Crawford WH (1983) Guidelines for the future of equine
locomotion research. Equine Vet J, 15:103-110.

84

Lewis GS, Preedy DF, Colbome GR (2002) Effect of variations in
morphological data on calculated mechanical energy cost in equine gait. Vet J.,
164(1):71-3.

Loslever P (1993) Error and data coding in the multi-dimensional analysis
of human movement signal. In: Proceedings of the Institute of Mechanical
Engineers, 207(2): 1 03-1 10.

Marey EJ (1873) In: La Machine Animale: Locomotion Tenestre et
Aerienne, 2nd edn Marey EJ (ed) Paris: Coll. Bibliotheque Science lntemationale,
Librairie Gemer Baillere et Cie, 145-186.

McLaughlin RM, Gaughan EM, Roush K, et al. (1996) Effects of subject
velocity on ground reaction force measurements and stance times in clinically
normal horses at walk and trot. Am J Vet Res, 57:7-11.

Merkens HW, Schamhardt HC, Hartman W, et al. (1985) Ground reaction
force patterns of Dutch Warmblood horses at normal walk. Equine Vet J, 18:207-
214.

Merkens HW, Schamhardt HC, Hartman W, et al. (1988) The use of
H(orse) INDEX: A method of analyzing the ground reaction force patterns of lame
and normal gaited horses at the walk. Equine Vet J, 20:29-36.

Merkens HW, Schamhardt HC, van Osch GJVM, et al. (1993) Ground
reaction force patterns of Dutch Warmblood horses at normal trot. Equine Vet J,
25:134-137.

Merkens HW, Schamhardt HC, vanosch ijm, at al. (1993) Ground
reaction force patterns of Dutch Warrnbloods at the center. Am J Vet Res,
54(5):670-674.

Merkens HW, Schamhardt HC (1988) Evaluation of equine locomotion
during different degrees of experimentally induced lameness. ll. Distribution of
ground reaction force patterns on the concurrently loaded limbs. Equine Vet J
(Suppl. 6):107-112.

Morales JL, Manchado M, Vivo J, et al. (1998) Angular kinematic patterns
of limbs in elite and riding horses at trot. Equine Vet J, 30(6):528-533.

Morris EA, Seeherrnan HJ (1987) Redistribution is ground reaction forces
in experimentally induced equine carpal lameness. Equine Exerc Physiol, 2:553-
563.

85

 

Mullineaux DR, Clayton HM (2003) Normalization of ground reaction
forces. In review.

Mullineaux DR, Wheat J (2002) Quantifying coordination in kinematic
data: a running example. In: Proceedings of the 20th lntemational Symposium on
Biomechanics in Sport (edited by K.E. Gianikellis). Caceres, Spain: University of
Extremadura, 515-8.

Mundy GD (1996) Regulating soundness and safety. J Equine Vet Sci, 16
(12):536-537.

Muybridge E (1887) Muybridge’s Complete Human and Animal
Locomotion. Vol 3 (Republication of Animal Locomotion). New York: Dover
Publications.

Newell KM, Corcos DM (1993) Issues in variability and motor control. In:
Variability and Motor Control. Champaign: Human Kinetics Publishers, 1-12.

Nigg BM, Walter H (1999) Biomechanics of the musculo-skeletal system
2"d ed. Chichester: John Wiley & Sons.

Nixon JA (1999) The Phalanges and Metacarpometatarsophalangeal
Joint. In: Auer JA, Stick JA. (eds) Equine Surgery 2" ed. Philadelphia: WB
Saunders, 792-808.

Noble ME & Bahrick HP (1956) Response generalization as a function of
intertask response similarity. J Exp Psycho/09% 51:405-412.

Palmer SE (1986) Prevalence of carpal fractures in Thoroughbreds and
Standardbred racehorses. J Am Vet Assoc, 188:1171-1173.

Peham C, Licka T, Girtler D, Scheidl M (2001) The influence of lameness
on equine stride length consistency. Vet J, 162 (2)2153-157.

Peham C, Licka T, Mayr A, et al. (1988) Speed dependency of motion
pattern consistency. J Biomech, 31(9):769-772.

Peloso JG, Mundy GD, Cohen ND (1994) Prevalence of and factors
associated with musculoskeletal racing injuries of Thoroughbreds. J Am Vet Med
Assoc, 204(4):620-6.

86

Peloso JG, Stick JA, Soutaslittle RW, et al. (1993) Computer-assisted 3-
dimensional gait analysis of amphotericin-induced carpal lameness in horses.
Am J Vet Res, 54(9):1535-1543.

Perry J (1992) Gait Analysis Systems. In: Gait Analysis: Normal and
pathological function. Thorafare: SLACK Incorporated, 349-443.

Pourcelot P, Degueurce C, Audigie F et al. (1997) Kinematic analysis of
the locomotion symmetry of sound horses at a slow trot. Equine Vet J, (Suppl.
23):93-96.

Pratt GW, O’Connor JT (1976) Force plate studies of equine
biomechanics. Am J Vet Res, 37:1251-1255.

Provins KA (1957) Sensory factors in the voluntary application of
pressure. Quarterly Journal of experimental Psychology, 9:28-41 .

Quddus MA, Kingsburry HB, Rooney JR (1978) A force and motion study
of the foreleg of a Standardbred trotter. J Equine Med Surg, 2:233-242.

Rabuffetti M, Frigo C (2001) Ground reaction: intrinsic and extrinsic
variability assessment and related method for artifact treatment. J Biomech,
34:363-370.

Ratzlaff MH (1989) Quantitative methods for the analysis of equine
locomotion and their applications to other species. Am Zool, 29(1):267-285.

Ratzlaff MH, Grant BD, Adrian M (1982) Quantitative evaluation of equine
carpal lameness. J Equine Vet Sci, 2:78-88.

Ratzlaff MH, Hyde ML, Grant BD, et al. (1990) Measurements of vertical
forces and temporal components of the strides of horses using instrumented
shoes. J Equine Vet Sci, 10:23-35.

Reboussin DM, Morgan TM (1996) Statistical consideration in the use and
analysis of single-subject designs. Med Sci Sports Exerc, 28:639-644.

Reinschmidt C, Nigg BM (1995) Influence of heel height on ankle joint
moments in running. Medicine and Science in Sports and Exercise, 27:410-416.

87

Riemersma DJ, Schamhardt HC, Hartman W, Lammertink JLMA (1988)
Kinetics and kinematics of the equine hind limb: in vivo tendon loads and force
plate measurements in ponies. Am J Vet Res, 49:1344-1352.

Riggs CM (2002) Fractures — A preventable Hazard of Racing
Thoroughbreds? Vet J, 163:19-29.

Robert C, Valette JP, Pourcelot P, Audigie F, Denoix JM (2002) Effects of
trotting speed on muscle activity and kinematics in saddlehorses. Equine Vet J
(Suppl), 34:295-301.

Ross MW (2003) Lameness in the horse: basic facts before starting. In:
Diagnosis and Management of Lameness in the Horse. Ross MW, Dyson SJ
(eds). St Louis: Saunders, 20-22.

Rossdale PD, Hopes R, Digby NJW (1985) Epidemiological-study of
wastage among racehorses 1982 and 1983. Vet Rec, 116(3):66-69.

Rozendal RH, Heerkens YF, van lngen Schenau GJ, et al. (1985) Vector
diagrams in the evaluation of human gait. Arch Phys Med Rehabil, 66:682-686.

Schamhardt HC, van den Bogert AJ, Hartman W (1993) Measurement
techniques in animal locomotion analysis. Acta Anat, 146:123-129.

Schamhardt HC (1996) Quantitative analysis of equine locomotion. In:
Ossenkopp, K-P, Kavakiers, M. and Sanberg, PR (eds) Measuring movement
and Locomotion: from invertebrates to Humans, Springer-Verlag, Heidlberg, 189-
211.

Seeherrnan HJ (1999) Lameness Evaluation. In: Auer JA, Stick JA. (eds)
Equine Surgery 2"d ed. Philadelphia: WB Saunders, 611-620.

Silver CA, Brown PN, Goodship AE (1983) Biomechanical assessment of
locomotor performance in the horse. Equine Vet J, (Suppl. 1):23-35.

Speirs VC (1994) Lameness: approaches to therapy and rehabilitation. In:
Hodgson DR and Rose RJ (eds) The athletic horse. Philadelphia: WB Saunders,
343-370.

Stallard J (2000) Lower limb orthotics. In: Clinical Biomechanics.
Philadelphia: Churchill Livingstone, 239-265.

88

Stashak TS (2002) Examination of lameness In: Adams: Lameness in
horses 5th ed. Philadelphia: Lippincott Williams & Wilkins, 113-116.

Stover SM, Johnson BJ, Daft BM, et al. (1992) An association between
complete and incomplete stress-fractures of the humerus in racehorses. Equine
Vet J, 24(4):260-263.

Swanson TD (1984) Guide for Veterinary Service and Judging of
Equestrian Events 3rd ed. Golden, CO: American Association of Equine
Practitioners, 19.

Theyse LFH, Hazewinkel HAW, van den Brom WE (2000) Force plate
analyses before and after surgical treatment of unilateral fragmented coronoid
process. Vet Comp Orthopaed, 13(3):135-140.

Ueda Y, Yoshida K, Oikawa M (1993) Analyses of race accident
conditions throughout use of patrol video. J Equine Vet Sci, 13:707-710.

Ueda Y (1991) Preventing accidents to racehorses: studies and measures
taken by the Japan Racing Association. In: Report pf the Committee on the
Prevention of Accidents to Racehorses, Japan Racing Association. Tokyo:
Equine Research lnstitute,1-16.

USDA (2001) National economic cost of equine lameness, colic and
equine protozoal myeloencephalitis (EPM) in the United States. In: Info Sheet
Veterinary Services. Ed: United States Department of Agriculture. Animal and
Plant Health Inspection Service, Colorado.

van Dixhoom ID, Meershoek LS, Huiskes R, Schamhardt HC (2002) A
description of the motion of the navicular bone during in vitro vertical loading of
the equine forelimb. Equine Vet J, 34(6):594-7.

van Emmerick REA, Wagenaar RC (1996) Effects of walking velocity on
relative phase dynamics in the trunk in human walking. J Biomech, 29:1175-
1184.

van Emmerik REA, Wagenaar RC, Winogrodzka A, Wolters EC (1999)
Identification of axial rigidity during locomotion in Parkinson disease. Archives of
Physical Medicine and Rehabilitation, 80: 1 86-1 91 .

Weeren PR van, Bogert AJ van den, Bameveld A (1990) Quantification of
skin displacement in trotting horse. Equine Vet J, (Suppl. 9):101-109.

89

Weeren PR van, Bogert AJ van den, Bameveld A (1992) Correction model
for skin displacement in equine kinematic gait analysis. J Equine Vet Sci, 12:178-
192.

Weeren PR van, Bogert AJ van den, Bameveld A (1988) Quantification of

skin displacement near the carpal, tarsal and fetlock joints of the walking horse.
Equine Vet J, 20(3):203-208.

Weeren PR van, Brama PAJ (2001) Physiology and pathology of the
equine joint. Pferdeheilkunde, 17(4):307.

Weeren van PR, Bogert AJ van den, Back W, et al. (1993) Kinematics of
Standardbred trotters measured at 6, 7, 8, and 9 m/s on a treadmill before and
after 5 months of training. Acta Anat (Basel), 146:154-161.

West GP (1984) Black’s Veterinary Dictionary. London: A & C Black, 442-
444.

Wheat J, Mullieaux DR, Bartlett RM, et al. (2002) Quantifying variability in
coordination during running. In: Proceedings of the 20th lntemational Symposium
on Biomechanics in Sport (edited by K.E. Gianikellis). Céceres, Spain: University
of Extremadura.

Whitall J, Getchell N (1996) Multilimb coordination patterns in
simultaneous, dissimilar upper and lower limb tasks. Hum Movement Sci
15(1):129-155.

White NA (2002) Fractures of the Equine Forelimb. Equine Surgery and
Medicine1 ,VM8695.
http://education.vetmed.vt.edu/CurriculumNM8695/naw/fracture/ForelimeX.pdf.

Whiting WC, Zemicke RF (1998). Mechanisms of injury. In: Biomechanics
of musculoskeletal injury. Champaign: Human Kinetics, 113-135.

Williams GE, Silverman W, Wilson AM, et al. (1999) Disease-specific
changes in equine ground reaction force data documented by use of principal
component analysis. Am J Vet Res, 60:549-555.

Wilson AM, McGuigan MP, Fouracre L, et al. (2001) The force and contact

stress on the navicular bone during trot locomotion in sound horses and horses
with navicular disease. Equine Vet J 33(2):159-165.

9O

 

Wilson DA, Keegan KG (1995) Pathophysiology and diagnosis of
musculoskeletal disease. In: Kobluck CN, Ames TR and Geor RJ (eds), The
horse, Disease & Clinical Management. Philadelphia: WB Saunders, 607-658.

Wittmann F (1931) Die Untersuchung auf Lahmheit. In: Wittmann F.
Chirurgische Diagnostik des Pferdes. Stuttgart: Ferdinand Enke Verlag, 140-149.

Wyn-Jones G (1988) The diagnosis of the causes of lameness. In: Equine
Lameness. Oxford: Blackwell Scientific Publications, 1-22.

Yamada N (1995) Nature of variability in rhythmical movement. Hum
Movement Sci, 14(3):371-384.

91

 

 

- . ' llllllllllilllllll'lllllllllllllllllllllllll'lll
, 3 1293 02504 7774