..
. L"

n»
.A.‘ .1 5..
.0 '1\r v.
u u. u 1:

“94
w

1..
”m
.a».

“n. v

5.. 4»
MW.
«ﬁ»..

5 -M .
mu

m...
.

F.-. -‘ ,
1:52:31
W

M...

u-w

 

- 12;...
”v:
’ﬁsqo

p—IR-

:‘ n1
!‘ I" q
33%

. «_¢
33‘“in

g.“ .
"i331“.
3.3“}.
[e
\H
h‘ "-
KM.
,559 ‘1'
x.

A

?'..s'

J. [:3
i2§§5i§§zg
' " 34:.

   

   

 

 

’ 1. ‘ .
i‘ ¢.,
*5. 13'}; ‘ .
a A ~ v ‘ :1. ‘
EH4; v: 3;. .‘E‘ﬁilqﬁiyi
v‘i ‘xag’vz‘ ,1 '
‘5‘! “3533’
' 9i ‘3‘;- '; ~2 i; '23:
T: 7‘6“§1t:‘§’- 1: 1 (1:43”:
.Q.\ [96! 2* 14‘5”“; I v
“‘3: “ ﬁg. 5;; '
3 WK 3-,? f}; .u
' , c ->-.
'1‘ zi'ﬁii“
D341“ .2: 3-:

3."
'ﬁ 0;
~37:
-NA .‘ .4—

a»,
.u.‘
,a ‘

my... «a
w

 

3
' I
V; t ’3'" pp
M'ﬁ’ﬁ‘“ 4 3‘13 4 . 139.931, -.. , 9'}
~ .ﬁsfi’éwigghéﬁﬁi . 9.2 i‘a»ﬁ§5’asﬁi‘?5?zf. “
.Lsiiéﬁiwr 'égsﬂfﬁsm» :}\ ‘21:”th 3,3; ' i. .
. ‘ ‘_ ‘42:? > _ F”

’
.o~.
.. .. 4.,
{‘4
:2".

 

.w...

 

31'

7
an
LL 4»

   

3‘ .
' a
| , 4 i -Q .3‘
. :9 Q- l h h ‘ 4‘. $33; , ,
’f‘ i (it zkizgkﬁ ‘t i' 1.513%"; 123’?*71>;’:E3:‘ ‘8
1.“, -' '2"5. ‘ _-. ~ 2-,.
gag; 23‘" “3;? ‘it‘bgéil fix? {4 v
a, - .up ”is? . .
x. rim" \

J! l
t‘:

THESE

sm LI IBRARIES

Hlll

MICHIG

l \lllllllllllll ll“ llllll \lml

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l

This is to certify that the

thesis entitled

MAGNIFICATION IN DIGITAL RADIOGRAPHY
AND DETERMINATION OF VESSEL DIAMETER
IN ANGIOGRAPHY OF THE EQUINE DISTAL LIMB

presented by

DIANA SUE ROSENSTEIN

has been accepted towards fulfillment

of the requirements for

MASTER OF SCIENCE
degree in

 

 

LARGE ANIMAL CLINICAL SCIENCES

yea/V Mb cal/92

Major professor

Date APRIL 26, 1996

 

0-7639 MS U is an Afﬁrmative Action/Equal Opportunity Institution

 

 

LRBRARY
Michmm atate

University

 

 

 

PLACE II RETURN BOXto removethte checkoutrom your record.
TO AVOID FINES return on or before dete due.

DATE DUE DATE DUE DATE DUE

      
 

 

 

 

 

 

 

 

 

 

 

 

 

l
l

Er- ﬂ:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU le An Afﬁrmative Adlai/Equal Opportunlty lnetituion
Mum-m

MAGNIFICATION IN DIGITAL RADIOGRAPHY AND DETERMINATION OF
VESSEL DIAMETER IN ANGIOGRAPHY OF THE EQUINE DISTAL LIMB

By

Diana Sue Rosenstein

ATHESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTEROF SCIENCE

Department of Large Animal Clinical Sciences

1996

ABSTRACT

MAGNIFICATION IN DIGITAL RADIOGRAPHY AND DETERMINATION OF
VESSEL DIAMETER IN ANGIOGRAPHY OF THE EQUINE DISTAL LIMB

By

Diana Sue Rosenstein

The size of an object may be determined from its radiographic image and
the magnification factor of the imaging system. Magniﬁcation factors of a digital
radiographic system were quantified by comparison of a metallic marker to its
image and by calculation of geometric, electronic and photographic magnification
factors from the equipment specifications. Angiography was performed in six in-
vitro equine feet, in the standing and lateral recumbent positions, for identification
and measurement of the digital arterial vessels. These measurements were
corrected for magnification to determine actual vessel diameters. Focal-film
distance, image intensifier mode of function and printer format were significant
predictors of variability in magnification factors. Total magnification factors
determined by the two methods differed due to widened penumbra with increased
magnification. Digital arterial diameters ranged from 0.71 mm to 2.54 mm and
were consistently larger in lateral recumbency, possibly due to vascular reactions

and pressure changes between the two positions.

DEDICATION

This manuscript is dedicated to my family who taught me to pursue my
interests with devotion and to my teacher and friend, John ”Doe” Walters, who

encouraged my intrigue for physics.

ACKNOWLEDGMENTS

I want to express my gratitude to my graduate committee members for their
guidance and patience in this endeavor: Drs. R. Bowker, N.E. Robinson, 1. Stick
and R. Stickle. I also appreciate the technical assistance and support that was

generously given by V. Hoelzer-Maddox and P. Ocello.

iv

TABLE OF CONTENTS

 

 

 

 

LIST OF TABLES ................................................................................................................. vii
LIST OF FIGURES ................................................................................................................. ix
LIST OF ABBREVIATIONS ............................................................................................. xiv
INTRODUCTION .................................................................................................................. 1
Radiography--- - .................................................... 1
Characteristics of the radiographic image ............................................................ 8
Angiography--- - - -- - -- - - ....... - - ............. 2
LITERATURE REVIEW .................................................................. . ................................... 29
Image magnification ............................................................................................... 29
Angiography ............................................................................................................ 30
Statement of hypothesis ......................................................................................... 43
MATERIALS AND METHODS ........................................................................................ 45
Magnification factors ......................................................................... 45
Limb perfusion and angiography ......................................................................... 51
RESULTS-- ................................................................................... 59
Magnification factors .............................................................................................. 59
Angiography of the equine distal limb ................................................................ 66
DISCUSSION ........................................................................................................................ 73
Magnification ........................................................................................................... 73
Angiography of the equine distal limb ................................................................ 78

V

CONCLUSION ..................................................................................................................... 83

APPENDICES ....................................................................................................................... 86
Appendix A .............................................................................................................. 86
Appendix B ............................................................................................................... 87
Appendix C ............................................................................................................ 100
Appendix D ............................................................................................................ 110

LIST OF REFERENCES ..................................................................................................... 117

vi

Table 1

Table 2

Table 3

Table 4

Table 5

Table 6

Table 7

Table 8

Table 9

Table 10

LIST OF TABLES

Following euthanasia, the right thoracic limb was removed on each of six
horses.

The palmar digital artery and its major branches to the foot were identified on
digital angiograms.

Total magniﬁcation factors ('I'MFm) were determined by measurement of a
metallic marker and its radiographic image.

Total magniﬁcation factors (TMFc) were calculated for the imaging system.

Arterial vessels were identified on angiograms of six equine feet.

Arterial vessels were measured on digital angiograms in the palmarodorsal
(PD) and lateromedial (1M) views and vessel size was corrected for

magniﬁcation.

A metallic marker was radiographed at several focal-ﬁlm distances (FFD), using
three different image intensiﬁer modes (II mode) and printed at four formats.
Each image was measured three times with calipers.

A linear function, defined by linear regression analysis, was used to calculate
ﬁtted values for TMFm at each focal-film distance (FFD), image intensiﬁer tube
mode (II mode) and printer format.

Magniﬁcation factors were calculated for individual components of the digital
radiographic imaging system, at each focal-ﬁlm distance, image intensiﬁer
mode and printer format.

There was some variation in the difference between measured and calculated
total magniﬁcation factors (delta) between the radiographic technique variables.

vii

51

56

60

66

71

87

95

100

101

Table 11 Arterial vessels of six in—vitro equine feet were measured on digital angiograms 110
(image size) in the palmarodorsal (PD) and corrected for magniﬁcation to
determine the diameter of each vessel (vessel size).

viii

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

LIST OF FIGURES

Electrons strike the target causing the release of xrays from the xray tube.

The image intensiﬁer tube converts xray energy to electrons and then light
photons.

The video camera records the light signal from the image intensiﬁer tube and
transmits the image to the digital image processor. The image may then be
viewed on a television monitor or printed by the laser imager. '

The triangles formed by the focal spot, object (a) and film (A) by the xray beam
(b, B, c, C) illustrate the geometrical relationship between the object and its
image.

As focal-object distance decreases, image size increases (A) and as focal-ﬁlm
distance increases, image size increases (B).

Xrays may originate from any random point in the target and this range causes
indistinct margins to the radiographic image called penumbra (P).

One light source casts a sharper image of a box (A) than the light from several
sources (B).

Electrons released at the input screen are directed towards the output screen by
electrostatic focusing lenses.

The image intensiﬁer tube functions in the 9 inch mode (A), 6 inch mode (B) and
4.5 inch mode (C) to cause electronic magniﬁcation by changing the focal point
of the electron stream.

10

11

12

12

14

16

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Iightfromthelasersourceisdirectedtotheﬁlmbyaseriesoflensesand
photographic magniﬁcation is determined by focal length of the collimator lens
(A) and focal length of the camera lens (B).

In a magniﬁed image (A) the individual dots are discemable, while a miniﬁed
image (B) blends together these dots, creating a sharper appearance.

Edge absorption causes image unsharpness due to the gradual attenuation of
xrays at the periphery of a structure. This effect is related to the shape of an
object and is least apparent with a conical object (A), moderate with a cubic
object (B) and most apparent with a spherical object (C).

Differences between the common radiopaque contrast media are illustrated by
their structural formulas. The ionic compormds (A) dissociate in solution
resulting in a higher osmolarity than the non-ionic compounds (B,C). The
monoacid dimer (D) is ionic, but the anion has six iodine atoms.

Digital radiographic images of a metallic marker were produced with the
marker on the patient table.

The vasculature was perfused with an oxygenated (95% 02/ 5% C02), heated
(37°C) Krebs Henseleit solution at 100 mm Hg arterial perfusion pressure.

The palmarodorsal study was performed with the limb in a standing position,
FFD=87.5 cm, FOD=49.5 cm and object-ﬁlm distance=38.0 cm.

The lateromedial study was performed with the limb in lateral recumbency,
FFD=74.S cm, FOD=49.5 cm and object-ﬁlm distance=25.0 cm.

The palmar digital artery was measured at three locations: 1) 1.0 cm proximal to
the origin of the bulbar a., 2) 1.0 cm proximal to the origin of the dorsal
phalangeal a. and 3) at the entrance to the solar canal.

The total magniﬁcation factors (TMFm) increased as FFD increased.

As II mode increased, the total magniﬁcation factors (TMFm) decreased
steadily.

The total magniﬁcation factors (TMFm) decreased as the number of images per
sheet of ﬁlm (format) increased.

18

20

28

47

53

55

57

61

61

61

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

As FFD increased, the difference between TMFm and MC (delta) remained

As 11 mode increased, the difference between M111 and TNti (delta)
decreased.

As printer format increased, the difference between 'I'MFm and TNch (delta)
decreased.

The palmar digital artery and its major arterial branches to the equine foot were
visible on digital angiograms in the palmarodorsal view (A) and diagrammed
as an atlas of these vessels (B).

The palmar digital artery and its major arterial branches to the equine foot were
visible on digital angiograms in the lateromedial view (A) and diagrammed as
an atlas of these vessels (B).

Dorsalbranchesoftheterminalarchand distalbranches ofthedorsal
phalangeal artery supply a vascular bed at the dorsal aspect of the coronary

region.

Minor variations were present in the pattern of the terminal arch and solar
branches, as seen on the palmarodorsal view.

Arterialvesselsoftheequinefootweremeasuredondigitalangiograms ofsix
feet and corrected for image magnification in the palmarodorsal (PD) and
lateromedial (LM) views.

Penumbra contributed to image magnification, but due to the ill-deﬁned,
gradually fading margins, it was not completely included in the measurements.

The magnitude of penumbra depends on the relative size of the object and the
focal spot. Penumbra (P) is small when the object is larger than the focal spot
(A) and penumbraislargerthantheumbra (U)whentheobjectissmallerthan
the focal spot (B).

Fitted values for total magniﬁcation factors (TMFm), averaged over all 11 modes
and formats, increased gradually with increasing FFD.

67

69

70

74

97

Figure 33

Figure 34

Figure 35

Figure 36

Figure 37

Figure 38

Figure 39

Figure 40

Figure 41

Figure 42

Figure 43

Figure 44

Fitted values for total magnification factors (TMFm), averaged over all formats,
were higher for smaller 11 modes, but as FFD increased, the trend of increasing
magniﬁcation factors was the same for each 11 mode.

Fitted values for total magniﬁcation factors (TMFm), averaged over all II modes,
were higher for formats with fewer images per sheet of film, but as FFD
increased, the trend of increasing magnification factors was the same for each
format. -

Fitted values for total magniﬁcation factors (TMFm), averaged over all FFD's
and formats, decreased steadily with increasing II mode.

Fitted values for total magnification factors (TMFm), averaged over all FFD’s,
were higher for formats with fewer images per sheet of ﬁlm. As 11 mode
increased, ﬁtted TMFm decreased at every format.

Fitted values for total magniﬁcation factors (TMFm), averaged over all formats,
were higher for longer FFD's, but as 11 mode increased, the trend of decreasing
magniﬁcation was the same for each FFD.

Fitted values for total magniﬁcation factors (TMFm), averaged over all FFD's
and II modes, decreased as format increased.

Fitted values for total magnification factors (TMFm), averaged over all II modes,
were higher for longer FFD's, but as format increased, the trend of decreasing
magnification factors was the same for each FFD.

Fitted values for total magniﬁcation factors (TMFm), averaged over all FFD's,
decreased as the number of images per sheet of ﬁlm (format) increased.

Fitted delta (the difference between measured and calculated magniﬁcation
factors), averaged over all 11 modes and formats, was the same at every FFD.

Fitted delta, averaged over all formats, was highest at the 4.5 inch 11 mode.
Fitted delta, averaged over all 11 modes, was highest at the 1:1 format.

Fitted delta, averaged over all formats and FFD's, was greatest at the 4.5 inch 11
mode.

97

97

98

98

98

99

107

107

107

108

Figure 45

Figure 46

Figure 47

Figure 48

Figure 49

Fitted delta, averaged over all FFD's, was highest at the 1:1 format.

Fitted delta, averaged over all formats, was the same at every FFD.

Fitted delta, averaged over all 11 modes and FFD's, was greatest at the 1:1
format.

Fitted delta, averaged over all II modes, was the same at every FFD.

Fitted delta, averaged over all FFD's, was highest at the 4.5 inch 11 mode.

108

108

109

109

DR

FOD
format
11 mode

kVp

PD
ECG
TMFc

TMFm

LIST OF ABBREVIATIONS

artery

digital radiography

focal-ﬁlm distance

focal-object distance

laser imager (printer) format

image intensiﬁer tube mode of function
kilovolts peak

lateromedial

milliamperage

palmarodorsal

electrocardiogram

calculated total magniﬁcation factor

measured total magniﬁcation factor

xiv

INTRODUCTION

RADIOGRAPHY
Physics of radiography

Radiography is a commonly performed diagnostic procedure in veterinary
medicine that provides visual images of structures within an animal's body, based
on xray attenuation by the tissues. The atomic number of the tissue and energy of
the xray determine the type of interaction that will occur between the patient and the
xray beam. When an xray ‘ is attenuated, either by absorption in the patient, or by
scatter to the environment, the corresponding region of the radiograph is
radiopaque and it appears clear, or white when the ﬁlm is viewed on a white light-
box. The area of a radiograph ﬂat is exposed to xrays and light from transmission
of the xrays through the patient is radiolucent and it appears as shades of gray to
black. Thus, a radiograph is a black and white image representing xray attenuation
characteristics of the subject, such as a patient’s body. Since the xray beam is
composed of thousands of individual xrays which vary in energy over a range and
an animal has a variety of internal structures, then the resulting radiographs contain

a spectrum of gray shades between white and black.

2
A conventional radiographic system includes an xray tube that emits x-

radiation, a cassette that contains two ﬂuorescent screens and the radiographic ﬁlm,
housed between these screens. Xrays that are transmitted through the patient strike
the cassette and interact with crystals in the screens, causing these crystals to emit
visible light. It is this light that exposes the radiographic ﬁlm to produce an image.

Digital radiography (DR) is a new development in radiology that incorporates
computer processing of data into image production to enhance the appearance of

radiographic images.

Digital radiography
Equipment

A basic explanation of xray production and digital radiographic equipment
illustrates the advantages of this sophisticated system. A radiograph is the product
of a series of complex events that uses energy from an electrical current to heat a
wire ﬁlament, causing the wire to release electrons. These electrons are attracted, at
an accelerating velocity across the tube, by the electrical potential difference between
the anode and cathode. The elecuons strike the target, causing rapid deceleration of
the electrons and the release of xrays and heat from the tube (Figure 1). The heat
generated in this reaction must be dissipated to avoid melting the target, while the
xrays are ﬁltered and collimated to form a useful beam. Xrays in this primary beam
are absorbed by the patient, scattered to the environment or transmitted through the

patient to create the latent xray image. In conventional radiography, this latent

3
image is a pattern of xrays ﬁrat excites crystals in the screens within ﬁre cassette
causing them to emit light to expose the adjacent ﬁlm. At this point digital

radiography differs from conventional radiography in its xray detection system and

 

 

image producﬁon.
"uncut FOCALSPOT mam
ANODB CATHODE

 

 

 

 

Figurel Elecﬁomsﬁikeﬁretargetcausingﬁrereleueofmysfromﬁremytube.

Transmitted xrays that form ﬁre latent image are detected by ﬁre image
intensiﬁer tube, which is positioned abOve' ﬁre patient. At ﬁre input screen of the
intensifier tube, xray energy is converted to electrons that are attracted across ﬁre

tube and focused at ﬁre output screen, where ﬁre energy is converted to light

4
photons (Figure 2). The image intensiﬁer tube is a sensiﬁve xray detector that

enhances ﬁre latent image signal by accelerating these electrons across a potenﬁal
difference, to release more light energy at ﬁre output screen ﬁran was present in ﬁre
initial latent inrage. The output image also appears brighter due to ﬁre relatively
smaller size of ﬁre output screen as compared to ﬁre input screen, which

concentrates ﬁre light photons over a smaller area.1

lllllll lllllll ”PM”

OUTPUT SCREEN

 

 

 

 

 

ELECTRONS

 

 

] INPUTSCREEN

Mttittttmm

Figure2 Timinragemtmrsiﬁermbemnvm‘tsmymergytoelecﬁomandﬁenﬁgtuphotons.

 

The intensiﬁer tube is optically linked to a television camera which records
ﬁre output light signal to produce a video signal. This message is transmitted to ﬁre
computer digital image processor for analog-to-digital conversion of ﬁre image data,

processing of ﬁre data and memory storage of ﬁre images. Data are ﬁren converted

5
back to an analog signal for viewing on a television monitor or transmitted to the

printer to produce a tangible radiograph. In ﬁre laser printer, ﬁris image signal
controls ﬁre intensity of a laser beam ﬁrat exposes radiographic ﬁlm. The exposed

ﬁlm is chemically developed in a processor to make ﬁre ﬁnal image visible and

 

 

 

 

 

 

 

 

 

 

 

permanent.
DIGITAL LASER
going“ was man
WK K (PRINTER)
MAGS
mm m
1113! MONITOR

 

 

 

 

 

 

 

/ PATIENTTABLE /
@mm

Figure?) Davideommerammrdsﬁmhglusigrulﬁomﬁremgemteraiﬁermbemduammitsﬁe
imagetoﬁredigitalimageprooessor. Theimagemayﬁrenbeviewedonatelevisionmonitoror
prirrtedbytheberimager.

Adomrtages afﬁne DR system

The principles of xray production and ﬁlm exposure are similar in
conventional and digital radiography, however ﬁre modiﬁcations present in ﬁre DR
system increase ﬁre efficiency of xray detection, incorporate computer manipulation
of ﬁre image data and allow for instantaneous viewing of ﬁre images on a monitor.
Characteristics of each component in ﬁre DR system contribute to its enhanced

image quality.

6

The xray tube used in digital radiography is designed to dissipate heat
efﬁcienﬁy to avoid tube damage under a heavy work load, which increases ﬁre
ability of ﬁre system to produce rapid interval exposures and to repeat series of
exposures wiﬁrout prolonged waiting periods for tube cooling.2 Turning of ﬁre
target by a rotating anode spreads ﬁre heat and wear, due to electron bombardment,
over a larger surface area to increase ﬁre heat loading capacity and extend ﬁre useful
life of ﬁre tube}!2 The shaft and bearings which attach to ﬁre rotating anode readily
conduct heat away from ﬁre target while an external blower cools ﬁre tube.2 The
target is constanﬁy bombarded by electrons during an exposure which causes
roughening or pitting of ﬁre target surface. The target material in ﬁre DR xray tube
is an alloy of tungsten (90%) and rhenium (10%) to increase resistance of ﬁre surface
to wear, over ﬁre conventional tungsten target.2

The intensiﬁer input screen is more efﬁcient at detecting transmitted xrays
and converting ﬁris energy to electrons, than ﬁre calcium tungstate screens in
conventional cassettesﬁ3 The crystals in ﬁre intensiﬁer tube input screens are cesium
iodide, which form long narrow crystals ﬁrat pack together more densely ﬁran
calcium tungstate, which increases ﬁre likelihood that a transmitted xray will
interact wiﬁr a crystal and inrproves image resolution by decreasing ﬁre dispersion
of light ﬁrat occurs wiﬁr large crystals, to create an electron signal ﬁrat more precisely
represents ﬁre latent xray image}!3 The intensiﬁer tube also ampliﬁes ﬁre signal
intensity by accelerating electrons across ﬁre tube. Faster electrons strike ﬁre output

screen with more energy, which causes ﬁre release of more light, ﬁrereby enhancing

7

image brightness. This enhancement of image brightness due to electron
acceleration between ﬁre input and output screens is called ﬂux gain.1 So ﬁre
intensifier screens in DR are more efficient at detecting xrays ﬁran conventional
screens and DR has ﬁre additional advantage of ﬂux gain to brighten ﬁre image.

The television camera and monitor allow for immediate viewing of ﬁre
image wiﬁrout chemical processing of ﬁlm. This visual system also links to a video
cassette recorder (VCR) to permit continuous recording of a dynamic study.” In
conventional radiography, each exposed ﬁlm must be chemically processed to create
a visible image and VCR capabiliﬁes are not available. The television chain,
composed of ﬁre camera and monitor, is ﬁre weakest link in ﬁre imaging system,
since it limits overall resolution” The standard system has 512 horizontal lines and
512 vertical lines and is called a 512x512 matrix; however, a newer television system
is now available wiﬁr a 1024x1024 matrix to improve image resoluﬁon. The televi-
sion monitor enables ﬁre radiographer to visualize images during ﬁre procedure for
immediate interpretation and to make adjustments in technique or patient
positioning as needed for a diagnostic study.

The digital image processor converts ﬁre electrical signal to a digital form, in
binary code, to process ﬁre image data. The computer may store 130 images in its
memory and recall these images for viewing and printing. The contrast and bright-
ness settings of each image may be changed by ﬁre windowing function, which
serves a similar purpose as fast, slow and par speed screens wiﬁr high and low

contrast ﬁlm in conventional radiography.” Since ﬁre windowing selections may be

8

altered after an exposure is complete, a single image may be viewed and printed at
several windows, to enhance various structures wiﬁrout repeating ﬁre radiation
exposure to ﬁre patient. The digital image processor has an image inversion
function which reverses ﬁre gray scale, such that radiolucent regions appear white
and radiopaque tissues appear black. An inverted image may enhance visualization _
of subﬁe radiographic ﬁndings, especially small structures in a ﬁrick body part.

The laser imager prints images on single-sided emulsion ﬁlm which has a
higher resolution ﬁran conventional double-sided emulsion radiographic film}:4
This improvement is due to ﬁre direct exposure of ﬁre ﬁlm to light from ﬁre laser
beam, instead of exposure to dispersed light from screens on two sides of ﬁre ﬁlm.1
These advantages of digital radiography, over conventional film-screen radiogra-
phy, make ﬁris newer imaging system well adapted for studies of small structures

ﬁrat require multiple, rapid interval exposures and high detail, high contrast images.

CHARACTERISTICS OF THE RADIOGRAPHIC IMAGE

The product of radiography is a two dimensional image that represents a
ﬁrree dimensional object, ﬁrerefore some discrepancy always exists between ﬁre
radiographic image and ﬁre object's true physical qualities. The image
characteristics ﬁrat represent ﬁrese differences are magniﬁcation, distortion and edge
unsharpness and each must be considered by ﬁre radiologist during image
interpretation.1 Some of ﬁre factors which contribute to these characteristics are

controlled by ﬁre radiologist and oﬁrers are inherent in ﬁre imaging modality. The

9
effects of ﬁrese characteristics may enhance or detract from image quality, so a

knowledge of ﬁrese factors enables ﬁre radiologist to manipulate ﬁre equipment for

maximum image quality.

Magniﬁcation
Geometric magniﬁcation

Magnification is deﬁned as ﬁre apparent enlargement of an object‘5 and three
types of magniﬁcation have been described: 1) geometric, 2) electronic and 3)
photographicmz8 In regards to radiography, geometric magnification is the most
commonly recognized form of image magnificationﬁm3 Xrays errritted from ﬁre
xray tube travel in straight lines from ﬁreir point of origin at ﬁre target, diverging
outward towards ﬁre patient. Geometric magnification is ﬁre enlarged appearance
of an image due to this diverging pattern of xray dispersion.” The focal spot is ﬁre
area of ﬁre target ﬁrat is bombarded by electrons to produce xrays. The distances
between this focal spot and ﬁre object and between the focal spot and ﬁre ﬁlm
determine ﬁre magnitude of geometric magniﬁcation. The triangles formed by ﬁre
focal spot, an object and ﬁre film illustrate the geometric relationship between the
object and its image (Figure 4).” Two triangles of ﬁre same shape but different size
are called similar triangles and ﬁre sides of ﬁrese triangles are proportional in ﬁre
ratio:

a/A = b/B = c/C = h/H.1

10
The altitudes of ﬁrese triangles represent ﬁre distances from ﬁre focal spot to

ﬁre object (11) and focal spot to ﬁre film (H). The object (a), produces an image (A) on
ﬁre ﬁlm and ﬁre sides of each triangle (b, c, B and C) represent ﬁre diverging xray

beam.

FOCAL SPOT

 

xrurv BEAM

 

 

 

 

 

Figure4 Thetrianglesfomredbyﬁrefocalspot,object(a)andfilm(A)bythexraybeam(b,B,c,C)
illustrate ﬁre geomeﬁical relationship betweentheobjectand its image.

The proportional relationship of ﬁre sides of similar triangles may be used to
calculate ﬁre size of an object from its image, or, vice versa, to predict ﬁre image size
of a known object. This geometric principle is used to explain how ﬁre focal spot to
object distance (FOD) and ﬁre focal spot to ﬁlm distance (FFD) affect image size
(Figure 5). As FOD decreases, ﬁre size of ﬁre image increases due to ﬁre divergence
of ﬁre xray beam (Figure 5A). As FFD increases, image size increases, also due to ﬁre

diverging pattern of ﬁre xray beam (Figure 5B). The term focal-ﬁlm distance comes

11
from conventional radiography, in which ﬁre ﬁlm is housed in ﬁre cassette.

HoWever, in digital radiography, ﬁre image intensiﬁer input screen detects ﬁre
transmitted xrays,butﬁreﬁlmisnotexposed until ﬁresignalisprocessedbyﬁre
computer and sent to ﬁre laser imager. In order to maintain consistency in ﬁre
terminology between ﬁre two imaging systems, ﬁre distance from ﬁre focal spot to

ﬁre input screen of ﬁre intensiﬁer tube is still called ﬁre focal-ﬁlm distance.

 

 

 

. A A A
F01) F017} (B) E
: FFD FFD'Z
' I
IMAGE ‘1
----- WAGE-"NJ wast-r V

Figures Asfomlobjedd'nhrmedeueases,imagesimuueases(A)mdasfoml-ﬁlmdism
irraeases,irrragesizeincreases(8).

Penumbra

In addition to FOD and FD, ﬁre focal spot size contributes a small amount to
image enlargement.1 The speciﬁc area of ﬁre target struck by electrons to release
xrays is ﬁre focal spot and each individual xray may be emitted from any random
point in ﬁre focal spot. This range of origins of ﬁre xrays causes an indistinct
appearance to ﬁre margins of an image, called ﬁre penumbra effect (Figure 6).”
Penumbra, also called edge gradient, is deﬁned as ﬁre region of partial illumination

ﬁr'at surrounds ﬁre complete shadow, or umbra.1 The penumbra appears as a region

12
of gradually dimming shades of gray at ﬁre margin of a structure's image. To illus-'

trate ﬁris phenomenon, one may inragine ﬁre light beam from one ﬂashlight, casting
a shadow of a box on a wall, resulting in a square wiﬁr sharp edges (Figure 7A). By
comparison, a box's shadow, cast by ﬁre light from four ﬂashlights, would have less

distinct edges due to ﬁre variety of angles from which ﬁre lights strike ﬁre box

(Figure 7B).

 

 

Figureé mesmayorigimteﬁemanymndompomtmﬁetargetandﬁusrangecausesmdisﬁnd
Myrntoﬁreradiographicimageoalledpenumbraw).

 

 

 

 

 

 

 

 

 

 

 

 

(A) (B)

Figure 7 One light source casts a sharper image of a box (A) than the light from several sources (B).

13

A larger focal spot will produce a wider penumbra, increase overall size of
ﬁre image and increase edge unsharpness, while a smaller focal spot will produce a
more narrow penumbra, resulting in a sharper image wiﬁr less effect on
magniﬁcation}9 In most xray tubes used in conventional radiography, ﬁre focal spot
is 1-2 mm, but ﬁre digital radiography unit has an xray tube wiﬁr two focal spot
sizes, 1.0 mm and 0.3 mm.2 The smaller focal spot (0.3 mm) is used in studies of
small structures to produce ﬁre sharpest possible images. The larger focal spot is
used to image larger structures, when ﬁne detail is less critical.”

Geometric magniﬁcation is created intentionally, in some studies, by using a
short FOD and long FFD, which is called ﬁre air gap techniqueﬁ10 The advantage of
ﬁris meﬁrod is enlargement of ﬁre image and absorption of scattered radiation in the
air gap, but penumbra degrades image quality, thereby lirrriting ﬁre applications of
ﬁris technique.10 These ﬁrree factors, FOD, FFD and focal spot size, affect every
radiograph and in conventional radiography ﬁreir impact is ﬁre predominant factor
ﬁrat determines image size. While geometric magniﬁcation does occur in digital
radiography, ﬁrere are oﬁrer factors to consider, which affect image size during
transmission of ﬁre image signal ﬁrrough the image intensiﬁer tube, computer and
laser printer ﬁrat must be recognized by ﬁre radiologist for proper image interpreta-

tion.

14

Electronic magniﬁaztion

There are two additional forms of image magnification ﬁrat occur in digital
radiography, namely electronic and photographic magrriﬁcaﬁon. Electronic
magnification is ﬁre alteration of image size by electronic manipulation of ﬁre image
signal, which occurs wiﬁrin ﬁre image intensiﬁer tube between ﬁre input and output
screens.‘ The electrons from a large input screen are focused to a smaller output
screen by electrostatic focusing lenses ﬁrat line ﬁre intensiﬁer tube (Figure 8).

OUTPUT SCREEN

d—b ................ ANODE
'. .- l. ..........
I ;; I VACUUM SEALED

-. ........ GLA$ TUBE

I .° -, | ........... ELECTROSTATIC
- - frocusmc; LENSES

 

 

 

 

 

 

 

 

L~_~.-¢EI.ECTRON STREAM

.....
e0.
.ueI
.....
,,,,,,,
.....
.ee
ce"
0..-

.' .vPI-IOTOCATHODE

V ~~mpurscmz~

Figure 8 Electrom released at ﬁre input screen are directed towards the output screen by electrostatic
focusinglemes.

 

 

 

The image is concentrated by projecting ﬁre signal on a smaller area, which
increases ﬁre density of electrons striking ﬁre output screen and releases light in a
more concentrated pattern, resulting in a smaller, brighter image wiﬁr greater
detail}3

The intensiﬁer tube also functions in three modes to produce additional

electronic magniﬁcation}7 The input screen is approximately 9 inches in diameter

15
(22 cnr) so these modes are called, by convention, 9 inch, 6 inch and 4.5 inch (Figure

9). In ﬁre 9 inch mode, all of ﬁre xrays detected at ﬁre input screen contribute to ﬁre
ﬁnal image (Figure 9A) and ﬁris all-inclusive relationship from ﬁre input to output
screens, in ﬁre 9 in. mode, does not cause a change in image size.1 In ﬁre 6 inch
mode, the electrons are focused at a point farﬁrer from ﬁre output screen by ﬁre
electrostatic focusing lenses, causing greater dispersion of ﬁre elect-ans, so ﬁrat
information from only ﬁre central 6 in. (66%) of ﬁre input screen contributes to ﬁre
image at ﬁre output side of ﬁre intensiﬁer tube (Figure 9B). The overall size of ﬁre
ﬁnal radiograph is ﬁre same, but contents wiﬁrin ﬁre image have been magniﬁed 1.5
times (9 inch/ 6 irrch=1.5).1 Finally, ﬁre 4.5 inch mode functions on ﬁris same
principle, uﬁlizing ﬁre central 50% of ﬁre input screen information to produce ﬁre

same size radiograph, so ﬁre signal is electronically magnified by a factor of 2
(Figure 9C).

16

 

(A) 9 men MODE

 

 

 

“FOCAL POINT

 

(B) 6 INCH MODE

 

 

 

 

 

 

_______ ”‘ -- (C)4.SINCHMODE _ ‘

 

 

Fisune9 Theimageintensiﬁertubefunctiominﬁre9indrmode(A),6inchmode(B)and4.5inch
mode (C) to cause electronic magniﬁartion by changing the focal point of ﬁre eledron stream.

17
The intensiﬁer tube mode of magniﬁcation may be easily interchanged

during a study, making it convenient for ﬁre radiologist to observe electronically
magniﬁed images on ﬁre television monitor. This ﬂexibility allows ﬁre radiologist
ﬁre opportunity to survey a large area, center on a selected region of interest, ﬁren
magnify ﬁre image by 1.5X or 2X. This technique is only available wiﬁr an intensifier
tube, which is a distinct advantage of digital radiography over conventional
radiography. The disadvantage of electronic magniﬁcation is decreased image
detail, since ﬁre same amount of image data is displayed over a larger area. To
maintain image brightness, in ﬁris electronic magnification mode, ﬁre computer
automatically increases ﬁre exposure technique, which increases radiation exposure
to ﬁre patient and adjacent personnel. In ﬁre 6 in. and 4.5 in. modes, ﬁre entire area
of ﬁre paﬁent ﬁrat is being exposed to ﬁre primary xray beam is not completely
visualized on ﬁre television monitor or on ﬁre ﬁnal radiograph. Therefore, personnel
restraining ﬁre patient may not realize ﬁreir close proximity to ﬁre primary beam,
resulting in excessive radiation exposure. In ﬁrese magniﬁed modes, ﬁre ﬁeld of
view wiﬁrin ﬁre patient is small, so only a limited area may be examined on each

image.

Photographic magnification
The ﬁrird type of image enlargement is photographic magniﬁcation, which
uses optical equipment, such as a camera lens, magnifying lens or microscope to

alter ﬁre size of an image.6 This type of magniﬁcation is well recognized in photog-

18
raphy wiﬁr ﬁre use of macro and zoom lenses to project light from subjects of

various sizes onto ﬁlm and to produce photographs of various sizes from a single
negative. A technique has been described for magnification radiography in which
ultra ﬁne detailed radiographs were produced of ﬁrin tissue specimens wiﬁr a
specialized xray tube, using low kVp, low mA and long exposure time settings. The
resulting radiographs were ﬁren photographed wiﬁr a zoom lens or viewed wiﬁr a
magnifying lensﬁnln digital radiography, photographic magnification occurs in ﬁre
laser inrager by altering ﬁre position of two lenses which control ﬁre laser beam!”-13

Light from ﬁre laser beam is aligned in parallel by ﬁre collimator lens, ﬁren

dispersed by ﬁre camera lens to expose ﬁre ﬁlm (Figure 10)."«13

LIGHT
COLLIMATOR APFFATURE

sue... i “€be 0 O
l O O
<—.—'> a——» O O

FigurelO Ugluﬁomﬁehsersourceisdhededmﬁeﬁlmbyasedesoflemesandphomgrapluc
magrriﬂmtionisdeterrninedbyfocallengthofﬁrecollimatorlens(A)andfocallengﬁrofﬁrecamera
lem(B). -

 

 

 

 

 

 

 

 

 

 

 

 

Photographic magniﬁcation is controlled by ﬁre focal lengﬁr of ﬁre camera
lens and ﬁre focal lengﬁr of ﬁre collimator lens. Moving ﬁre camera lens closer to ﬁre
ﬁlm creates a smaller image, while moving ﬁre camera lens furﬁrer away from ﬁre

ﬁlm produces a larger image.13 The .laser imaging printer is programmed to

19

produce images of various sizes, to ﬁt 1, 4, 6 or 12 circular images on one 14x17"
sheet of ﬁlrrr and ﬁrese formats are named 1:1, 4:1, 6:1 and 12:1, respectively.” The
printer format is selected by ﬁre operator to suit ﬁre specific needs of an examination.
This option is cost effective, since less ﬁlm is required to print multiple images and a
smaller volume of developing chemicals is used to process ﬁre ﬁlm.

These three types of magnification, geometric, electronic and photographic,
occur in digital radiography to create circular images on filnr ﬁrat may be smaller or
larger ﬁran ﬁre actual area of ﬁre patient wiﬁrin ﬁre xray ﬁeld. If ﬁre circular image is
smaller ﬁran ﬁre ﬁeld of view in ﬁre patient, ﬁren ﬁre image is miniﬁed, or magniﬁed
by a factor less ﬁran one. Of ﬁrese ﬁrree types of magnification, only electronic and
photographic magniﬁcation may have factors less ﬁran one, so miniﬁcation wiﬁrin
ﬁre radiographic system can only occur in digital radiography and not in
conventional radiography.

Mirriﬁcation of an image is beneﬁcial to image quality because ﬁre visual
information is condensed into a smaller area, increasing ﬁre density of information
per unit area. For example, if a newspaper photograph is enlarged, ﬁre individual
dots ﬁrat create ﬁre picture are discernible as separate spots of gray ink on white
paper and ﬁre subject appears less distinct (Figure 11A). But when ﬁre picture is
small, ﬁre dots blend togeﬁrer and ﬁre subject appears well deﬁned (Figure 11B).1
This logic applies to digital images where miniﬁed images appear sharper because a
discrete quantity of data is printed in a smaller area on ﬁre ﬁlm. In miniﬁed images,

ﬁre penumbra is smaller so ﬁre edges of structures appear sharper.

 

Figure 11 In a magniﬁed image (A) ﬁre individual dots are discernible, while a miniﬁed image (B)
ﬁrese dots, treating a sharper appearance.

Distortion

Distortion is ﬁre second characteristic of a radiograph that represents ﬁre
disparity between an object and its image. Distortion is ﬁre unequal magniﬁcation
of different parts of an object due to varying distances from ﬁre ﬁlm.” This detracts
from image quality due to ﬁre awkward, asymmetrical representation of ﬁre object.
To avoid distortion, an object should be aligned parallel to ﬁre patient table and
intensiﬁer tube such ﬁrat every part of ﬁre object is equidistant to ﬁre ﬁlm.1
However, positioning of irregularly shaped, ﬁrree dimensional patients usually
means ﬁrat internal organs cannot all be equidistant to ﬁre film. Interpretation of
distorted images requires comprehension of ﬁre anatomy and experience by ﬁre

radiologist.

21

Image unsharpness

Image unsharpness is ﬁre ﬁrird characteristic of radiographs ﬁrat changes ﬁre
appearance of an object. Unsharpness refers to ﬁre indisﬁnct margins of an object
and it is caused by ﬁrree factors, including, penumbra, patient or tube motion and
edge absorption.1 Use of a small focal spot will decrease penumbra, ﬁrus improving
image sharpness (Figure 7, p. 12). Patient restraint by physical or chemical means,
or use of a short exposure time, will decrease blurring from patient moﬁon. Xray
tube motion may occur wiﬁr a portable xray unit ﬁrat is not properly stabilized, but
ﬁris is not a problem in digital radiography, since ﬁre xray tube is ﬁxed in position.

Edge absorption is ﬁre gradual change in xray attenuation at ﬁre margins of
an object, associated wiﬁr ﬁre shape of a structure (Figure 12).” A spherical object
causes ﬁre greatest edge absorption because ﬁrere is always a gradual transition in
object ﬁrickness at ﬁre periphery.1 A cone shaped object causes ﬁre least edge
absorption since ﬁrere is an abrupt transition between ﬁre thick portion of ﬁre object
and ﬁre surrounding air and a cube causes a moderate amount of edge absorption,
between ﬁre sphere and cone. The shape of an organ cannot be altered to avoid edge
absorption, so ﬁre radiologist must learn to interpret images wiﬁr ﬁris characteristic.
These causes of magniﬁcation, distortion and unsharpness contribute to ﬁre quality

of a radiograph and are important to consider in image interpretation.

 

 

 

 

A)

Figure 12 Edge absorption causes image umharpnees due to ﬁre gradual attenuation of xrays at ﬁre
peripheryofastrudure. Th’maffedisrelatedtoﬁreshapeofanobjectandisleastapparentwiﬁra
corrimlobjed(A),rnoderatewiﬁracubic object (B) and most severewiﬁraspherical object (C).

ANGIOGRAPHY
Purpose and technique

Since blood vessels and blood are soft. tissues, composed mainly of water,
ﬁrey have ﬁre same xray attenuation characteristies as ﬁre surrounding muscles,
ligaments and tendons. To enhance visualization of ﬁre blood vessels, a radiopaque
solution may be injected into ﬁre vasculature immediately prior to radiography and
ﬁris procedure is called angiography.5 Vessels are evaluated on angiograms for ﬁreir
size, number and distribution wiﬁrin an organ, integrity of ﬁreir walls and
uniformity of lumirral diameter.‘5

Angiography of ﬁre heart may be performed following injection of a contrast
solution in a peripheral vein (non-selective cardiography) or following

caﬁreterization of speciﬁc cardiac chambers (selective cardiography). For ﬁre

23
exarrrination of oﬁrer blood vessels, a peripheral artery or vein is selected based on

ﬁre region of interest.

Contrast media
Systemic reactions to contrast media

Ideally, ﬁre only action of ﬁre contrast material is to attenuate radiation so as
to enhance visualization of ﬁre vessel lumen on ﬁre radiograph, wiﬁrout having any
biological effects.“ However, when ﬁrese radiopaque compounds are adrrrinistered
into ﬁre vascular systenr, a number of local and systemic reactions may occur, which
in turn, may affect ﬁre radiographic appearance of ﬁre vessels.“ These effects must
be considered in interpretation of angiographic ﬁndings, but are rarely mentioned in
ﬁre veterinary angiography literature.

Several body systems are commonly involved in contrast media induced
reactions including ﬁre neurological, respiratory, cardiovascular, gastrointestinal,
urinary and cutaneous systems?”14 Neurological reactions vary in severity from
mild ataxia, weakness and muscle twitching, to syncope, seizures, paralysis and
comaﬁtr" Reactions of ﬁre respiratory system include sneezing, coughing, dyspnea,
tachypnea, laryngospasm, cyanosis, pneumoﬁrorax, pulmonary edema, pleural
effusion and respiratory arrestlor“ Cardiovascular responses may be observed in ﬁre
electrocardiogram, such as, depression of ﬁre ST segment, lowering or inversion of
the T wave, premature ventricular beats and sinus arrhyﬁrmiasﬁ‘mv“ Vasodilation is

reported wiﬁr decreased vascular resistance, decreased blood pressure and

24
increased blood ﬂow, followed by a gradual return to normal pressuresﬂ‘ll‘t15

Gastrointestinal reactions commonly include vomiting, nausea and a metallic taste
sensation)“17 These symptoms reported in human medicine may be displayed in
ﬁre veterinary paﬁents as retching and licking of ﬁre lips. Oﬁrer gastrointesﬁnal
complications, associated wiﬁr abdominal aortic angiography are paralytic ileus,
mesenteric thrombosis and intestinal perforation.”15 Reactions of ﬁre renal system
vary from transient oliguria and albuminuria to hemorrhagic nephrosis, anuria and
acute renal failure.”«17 Cutaneous reactions in humans are often mild hypererrria
and warmﬁr sensation near ﬁre injection site, while urticaria, edema and gangrene
have also been reported.”14 The exact mechanisms of ﬁrese systemic reactions are
unknown, but several physical properties of ﬁre contrast media have been
implicated as ﬁre causal factors. Osmolality, iodine concentration, viscosity, pH and
iodine content of ﬁre various contrast solutions have been proposed as signiﬁcant
characteristics of ﬁre compounds which relate to contrast-induced reactions?”

These chemical properties were modiﬁed during ﬁre development of less toxic
contrast media in order to decrease ﬁre incidence and severity of contrast-induced

reactions.18

Evolution of contrast media
Inorganic and organic compounds
In 1896, only two monﬁrs after Wilhelm Roentgen announced ﬁre discovery

of xrays, ﬁre ﬁrst angiogram was performed on an amputated human hand ﬁrat

25
demonstrated ﬁre potential use of radiography for visualization of ﬁre arterial

vasculature. The original contrast media used on post-mortem specimens contained
lead or mercury to provide excellent radiographic contrast to illustrate ﬁre
anatomical structures, but ﬁrese compounds were not safe for use in live patients.“
In 1923, Sicard and Forestier used a bismuﬁr and oil suspension, Lipidol®‘, for
intravascular studies, which produced adequate radiographic contrast, but formed
globules in ﬁre vessels, causing pulmonary emboli and deaﬁr in experimental dogs.14
In 1927, an inorganic, sodium iodide compound was used for contrast radiography
of ﬁre kidney ﬁrat provided good image contrast, but it was too toxic for use in live
patients.“ By ﬁre end of ﬁris decade, an organic iodide compound, called
Selectan®b, was discovered, ﬁrat opaciﬁed ﬁre blood vessels and ﬁre collecting
system of ﬁre kidneys following an intravascular injection and was well tolerated by
ﬁre body.“ Several variations of organic contrast agents were developed over ﬁre
following twenty years wiﬁr ﬁre purpose of increasing radiographic contrast while

decreasing ﬁre incidence and severity of systemic reacﬁons.

Triiodin_a_ted ben_zoigcid derivatives: ionic and non-ionic compounds
There was litﬁe improvement in contrast media during ﬁris ﬁme, until ﬁre

1950's, when ﬁre organic, triiodinated benzoic acid derivatives were simultaneously

 

' Savage Laboratories, Inc., Houston, TX
" Schering AG, West Germany

26

discovered in ﬁre United States and Germany.14 The diatrizoates (Hypaque®°,
Renograﬁrr®d) and ioﬁralamate (Conray®e) are relatively safe organic contrast
agents ﬁrat are still used in modern angiography (Figure 13A). These compounds
dissociate in solution to an anion containing ﬁrree iodine atoms, and a cation, which
is usually sodium or meﬁrylglucamine (meglumine). These ionized particles create
a high osmolar compound ﬁrat is hypertonic to plasma, which may account for
many of ﬁre adverse effects of intravascular contrast media.”17 The most significant
advancement in contrast media evolution was Torsten Almén's discovery of ﬁre non-
ionic compounds.18 Metrizamide (Amipaque®f) was ﬁre ﬁrst generation of non-
ionic contrast media ﬁrat did not dissociate in solution, however, metrizamide was
expensive to produce and unstable during autoclave sterilization (Figure 13B .19
Iopamidol (Isovue®9, Niopam®") and iohexol (Omnipaque®i) contain ﬁrree iodine J
atoms per molecule, do not dissociate in solution and retain their stability ﬁrrough

steam sterilization (Figure 130.1“o onaglate (Hexabrix®j) achieves ﬁre same ratio

 

c Winﬁrrop Pharmaceuticals, Sterling Drug Inc., New York, NY

‘1 ER. Squibb 8: Sons, Inc., Princeton, N]

e Mallinckrodt, Inc., St. Louis, MO

‘ Nyegard 8: Co., Oslo, Norway

8 Squibb Diagnostics, Princeton, NJ

1‘ Bracco Industria Chimca Spa, Milan, Italy

‘Nyegard & Co., Oslo, Norway (now available through Nycomed, Inc., New York, NY)

t May and Baker, LTD., Essex, England

27

of iodine atoms to particles in solution (3:1) by linking togeﬁrer two triiodinated
benzene rings to form a monoacid dimer (Figure 13D). Only one cation is released
in solution and ﬁre anion contains six iodine atoms, so ﬁre ratio of iodine atoms to
particles in solution is 6:2, which is equivalent to 3:1.“3-20 Therefore, ﬁre second
generation of non-ionic compounds has a 50% lower osmolality than ﬁre ionic com-
pounds and produces ﬁre same radiopacity since it contains ﬁre same number of
iodine atoms per molecule. The non-ionic, low osmolar, contrast media have a
lower incidence rate of reactions, following angiographic proceduresﬁﬁ'z‘l21 Such
advances in contrast materials have improved ﬁre quality and safety of angiographic
studies.22

In human radiography, ﬁre beneﬁts of non-ionic, low osmolar contrast agents
have been recognized for years and clinically applied in high-risk patients.23 The
disadvantage of ﬁris new generation of contrast media is ﬁreir high cost, which often
linrits ﬁreir use in veterinary radiography to studies involving ﬁre brain and spinal
cord, in which ﬁre high osmolar, ionic agents are highly toxic to neural tissues.20
Since digital radiography is capable of producing and recording rapid interval
exposures, repeat injections are rarely necessary and ﬁre volume of contrast material
needed for a study is decreased, ﬁrereby reducing ﬁre cost and ﬁre likelihood of

contrast-induced reactions.15

28

ONe (SODIUM)
(A) or
OﬁNHCl-lgdigd-lphd-QOH (MEGLUMINE)
I I
NHCOCH, (D IATRIZOATE)

0f
OONHCt-l,(IOTHALAMATE)

METRIZAMIDE
COG-I

   

 

oouncmcnpri),
I
caprorroo Macao-rpm,
IOI’AMIDOL
W
comer WSW:

(D) 1 I I I
IOXAGLATE
3°! NHooa-wr-rco CONHCHa

Figure 13 Diﬁmences between ﬁre common radiopaque contrast media are illustrated by ﬁreir
strudural formulas. The ionic compounds (A) dissociate in solution resulting in a higher osmolarity
ﬁran the non-ionic compounds (B, C). The monoadd dimer (D) is ionic, but ﬁre anion has six iodine
atoms.

LITERATURE REVIEW

IMAGE MAGNIFICATION

Several reference textbooks provide an explanation of ﬁre geometry and
physical characterisﬁcs of radiography.”13 Based on ﬁrese principles, formulas have
been written ﬁrat account for geometric and electronic image magnification. A
maﬁrenratical formula has also been written ﬁrat incorporates ﬁre contribution of
penumbra to ﬁre size of a radiographic image. Unfortunately, since penumbra is
literally a ’gray zone’, each observer may determine ﬁre edge of an image to be at a
slighﬁy different location. Spears, et a], recorded angiographic studies of phantom-
model vessels on 35 mm ﬁlm ﬁren digitized ﬁrese images for computer analysis.
The computer program determined ﬁre vessel margins based on an algoriﬁrm ﬁrat
located points of increased radiopacity at speciﬁc spatial densiﬁes. A comparison of
ﬁre computer—derived vessel diameters to ﬁre known diameters showed ﬁrat ﬁre
only variable which affected accuracy of ﬁre computed data was ﬁre concentration of
contrast material wiﬁrin ﬁre vessel. The computer program was unable to locate ﬁre

lumirral edges of vessels less ﬁran 1 mm in diameter.”

30
Mensuration of structures, based on ﬁreir radiographic appearance has been

described by two meﬁrods in ﬁre veterinary literature. Linford, et al, reported on ﬁre
normal lengﬁr of ﬁre epiglottis in ﬁre Thoroughbred horse.25 To account for image
magnification, two radiopaque markers were positioned, one on each side of ﬁre
horses head. The magrriﬁcaﬁon factor at each side of ﬁre head was determined by
division of ﬁre actual marker size by ﬁre radiographic marker size. The epiglottis is
a midsagittal structure, so ﬁre magnificaﬁon factor at ﬁris level was determined by
ﬁre average of ﬁre magniﬁcation factors at each side. The second approach to
magnification is comparison of structures wiﬁrin ﬁre patient, included on ﬁre same
radiograph, as a ratio, such as pulmonary vessel size to rib widﬁr“ or heart diameter
to ﬁroracic cavity widﬁr.” These relationships are not uniformly accepted wiﬁrin ﬁre
literature.” No reports were found in ﬁre radiology literature ﬁrat compared
magnification factors based on ﬁre geomeﬁy of ﬁre imaging system to magniﬁcaﬁon
factors determined by measurement of ﬁre radiographic image. No quantitaﬁve
accounts have been made for electronic or photographic magniﬁcation, which are

inherent in digital radiography.

ANGIOGRAPHY
Angiowdiosmrhy

Soon after ﬁre discovery of xrays by W. Roentgen in 1895, ﬁre science of
radiography was used for anatomical studies and medical diagnostic procedures.

The original angiographic studies were performed on post mortem human and non-

31
human specimens, wiﬁr a variety of radiopaque substances.“ In 1896, Morton and

Hammer published a book on radiography, entiﬁed, The xray, or, photography of the
invisible and its value in surgery, in which ﬁrey wrote:

In teaching ﬁre anatomy of ﬁre blood vessels, ﬁre xray opens out a new

and feasible meﬁrod. The arteries and veins of dead bodies may be

injected wiﬁr a substance opaque to ﬁre xray and ﬁrus, ﬁreir

distribution may be more accurately followed ﬁran by any possible

dissection.29

Angiography has been used in veterinary medicine for identiﬁcation of
normal vascular anatomy and paﬁrology in many species. N on-selective
angiocardiography is a radiographic study of ﬁre heart, following a peripheral
venous injection of a contrast solution. Techniques for ﬁris procedure have been
described in ﬁre normal dog, can” and horse”1 wiﬁr an iodine based contrast
solution followed by serial radiographs. Wise, et aL used an ionic contrast media
and reported electrocardiographic arrhyﬁrrrrias and two fatalities following repeated
high dose injections in dogs.30 Carlsten also reported mild transient alterations in.
ﬁre ECG wiﬁr occasional ventricular premature contractions, following a peripheral
venous ionic contrast 9 media injection, in horses undergoing non-selective
angiocardiography.31 N 0 severe reactions were observed in Carlsten's study;
however, ﬁre advantages of ﬁre non-ionic compounds for lowering ﬁre likelihood of
adverse reactions were discussed, indicating ﬁre awareness of veterinary
radiologists to ﬁre concern of direct and systemic effects of ﬁrese ionic contrast
media.31 Non-selective angiocardiography has been reported for diagnosis of

congenital anomalies in dogs.32 The technique described in ﬁris paper suggested

32
ﬁrat any organic iodine contrast media may be used, at a dose of 0.5-1.0 ml/ kg. A

peripheral venous injection was made as a rapid bolus, followed by 4-6 radiographs
at 1-2 second intervals. A simple design for a radiographic cassette changer was
described so ﬁrat ﬁre procedure may be completed wiﬁrout sophisticated equipment.
Radiographs taken 45 seconds after ﬁre injection demonstrated contrast
enhancement of ﬁre right heart and pulmonary circulation and later ﬁlms showed
contrast material in ﬁre left heart. The auﬁrors acknowledged ﬁrat a slow bolus
injection caused dilution of ﬁre contrast and late enhancement of ﬁre right heart.
Superimposition of ﬁre left and right side structures made interpretation and
diagnosis of some anomalies more difﬁcult. Some ionic contrast media are more
viscous ﬁrerefore more difﬁcult to inject rapidly, ﬁran ﬁre non-ionic media.” Use of
a low viscosity, non-ionic compound would improve image quality by increasing
ﬁre concentration of contrast material in a rapid bolus and decreasing ﬁre problems
associated wiﬁr delayed right heart ﬁlling from ﬁre slow bolus.

Selective angiocardiography involves ﬁre placement of a caﬁreter into speciﬁc
cardiac chambers, permitting ﬁre contrast agents to be delivered at ﬁre target site of
interest, followed rapidly by serial radiography.” Fluoroscopy, a form of
radiography ﬁrat produces a dynarrric visual image, is recommended for caﬁreter
guidance to ﬁre target location. Most modern cardiac caﬁreterization imaging
equipment is a form of digital radiography wiﬁr specialized programs for cardiac
studies.3 The rapid interval images require use of an xray tube ﬁrat can wiﬁrstand

ﬁre heat production and be cooled rapidly to avoid tube damage. Digital

33
subtraction angiography, a speciﬁc function of digital radiography for cardiac

caﬁreterization studies, uses a mask image to subdue ﬁre images of background

structures so ﬁrat successive images will only show ﬁre injected contrast material.”

Peripheral mgiogmphy

Peripheral angiography has been used to demonstrate ﬁre arterial anatomy of
ﬁre abdominal aorta in ﬁre goat, dog, pig and rabbit.33 This study noted excellent
radiographic contrast wiﬁr a lead suspension contrast media for illustration of ﬁre
anatomy of ﬁre abdominal aorta, however, ﬁris technique cannot be used in live
patients to be recovered from ﬁre procedure, since ﬁre lead suspension obstructs
capillary beds.33 Singh, et a1, performed angiography using ioﬁralamate and lead
suspension on bovine tissue specimens to assess intestinal healing following ﬁrree
anastomotic procedures.“ Two images of ﬁssues injected wiﬁr ioﬁralamate, on ﬁre
ﬁrst day following surgery, demonstrated hypervascularity of ﬁre intestinal tissues
and vessel dilation, but later studies, performed on post-mortem tissues wiﬁr ﬁre
lead suspension, did not show ﬁris dilated appearance to ﬁre vessels. The tissue
hypervascularity seen on ﬁre earlier images may have been associated wiﬁr ﬁre
surgical procedure and early tissue healing; however, ﬁre vascular response to ﬁre
infusion of ﬁre ionic contrast material, ioﬁralamate, may have caused ﬁris vasodila-
tiorr.21 Singh, et al, also examined ﬁre vascular response to fracture healing in ﬁre ox
by serial angiography following various meﬁrods of fracture repairﬁE5 A pattern of

increased soft tissue vascularity and vasodilaﬁon was demonstrated wiﬁr

34
ioﬁralamate, at ﬁre fracture site for four weeks following stabilization of ﬁre

traumatized bone. Angiographic studies of ﬁre post-mortem specimens in ﬁris
study, using lead suspension, did not demonstrate ﬁris hypervascular pattern, which
was interpreted as regression of ﬁre vascular proliferation wiﬁr ﬁre healing process.
However, ﬁre lead suspension angiograms also failed to enhance major arteries to
ﬁre long bone. The disparity between ﬁre hypervascular pattern on ﬁre ioﬁralamate
angiograms and ﬁre hypovascular pattern on ﬁre lead suspension angiograms may
have been associated wiﬁr ﬁre healing process, as suggested by ﬁre auﬁrors, but ﬁre
affect of ﬁre contrast media should also have been considered.

Rendano, et al, performed repeated arteriographic studies in foals, to assess
ﬁre changes in ﬁre cranial mesenteric artery wiﬁr Strongylus vulgar-is parasitismﬁ‘S A
caﬁreter was passed, using ﬂuoroscopic guidance wiﬁr an image intensiﬁer tube,
from ﬁre common carotid artery to ﬁre abdominal aorta, ending just cranial to ﬁre
celiac artery. Thirty to forty milliliters of an ionic contrast media was used for ﬁre
angiogram and a single radiograph was taken one second after ﬁre injection. A
second dose of contrast media was administered, in order to take a second
radiograph, two seconds post-injection. The use of a digital radiography system
wiﬁr rapid interval exposures would have eliminated ﬁre need for a repeat injection,
which wOuld decrease ﬁre risks and expenses associated wiﬁr additional contrast
material administration. Vascular dilation, corrugation and ﬁrrombosis of ﬁre
cranial mesenteric, ileocolic and right colic arteries were observed in infected foals,

as well as dilations and irregularities in ﬁre hepatic, left gastric, caudal mesenteric,

35
jejunal and cecal arteries. Wiﬁr treatment, ﬁre vascular lesions regressed but ﬁrere

was some persistence of mildly dilated arteries on subsequent angiograms. The
authors attributed ﬁre persistenﬁy abnormal vascular patterns observed in ﬁris
study, including vascular occlusion and delayed blood ﬂow, to ﬁre parasitic
infection, because it was contrary to ﬁre vasodilation and increased blood flow
associated wiﬁr contrast media. However, ﬁre ionic contrast agents have been
shown to cause alterations in red blood cell morphology ﬁrat may lead to vascular
ﬁrrombosisl‘t21 and direct myocardial toxicity of ﬁre contrast media may cause
bradycardia, secondary to increased vagal tone.16 Therefore, ﬁre results described in
ﬁre study by Rendano, et a1, may still have been due to contrast-induced

hemodyrramic reactions.

Other imaging modalities: ultrasonography, MRI, nuclear scinligraphy

Several less invasive imaging modalities have been applied to ﬁre
examination of ﬁre cardiovascular system, including ultrasonography37‘“, Doppler
ultrasonography‘w, magnetic resonance imaging‘ﬁ- 47 and nuclear scintigraphy.“ ‘9
While each of ﬁrese modalities offers certain advantages, ﬁrey are limited in ﬁreir
examinations of ﬁre smaller peripheral vessels. In ultrasonography, an image is
created by ﬁre attenuation and reﬂection of sound waves at tissue interfaces.37 This
modality is effective for identification of large vessels; however, ﬁre resolution of
small structures is limited by ﬁre probe size for axial resolution and lengﬁr of a

sound wave packet for depﬁr resolution.1 Contrast sonography has been described

36
for examination of ﬁre equine heart and identification of cardiac shunting anomalies,

using saline, blood, 5% dextrose, C02, and indocyanine green for contrast agents.31
A mixture of C02 and blood provided ﬁre most sonographic contrast in ﬁre heart,
but ﬁris technique has not been applied to peripheral vessels.31 Transrectal sono-
graphy has been used to evaluate ﬁre cranial mesenteric and ileocolic arteries in
normal horses and horses infected wiﬁr S. vulgaris.“ 39 Ultrasonography has been
reported as an effective, relatively non-invasive meﬁrod for diagnosis of aorto-iliac
ﬁrrombosis in ﬁre horse for evaluation of large vessels, accessible via ﬁre
transabdominal or transrectal approach.” 41

The Doppler effect is an apparent change in ﬁre frequency of sound waves
reﬂected from ﬁre interface of a tissue ﬁrat is moving relative to ﬁre sound wave
receiver.42 Doppler ultrasonography is used for evaluation of laminar and turbulent
blood ﬂow ﬁrrough ﬁre heart and peripheral vessels:B Continuous and pulsed
Doppler produce a graphic image of ﬂow velocities and direction, while color ﬂow
Doppler adds ﬁre visual aid of color to various velocities and directions of ﬂow on a
real time, two dimensional image.1 A subcutaneous arteriovenous ﬁstula on ﬁre
thoracic body wall of a horse, was diagnosed wiﬁr Doppler sonography.“ Contrast
angiography was not performed in ﬁrat case because ﬁre originating vascular supply
to ﬁre lesion could not be idenﬁfied. The applications of ultrasonography are
limited by ﬁre inability of sound waves to effectively penetrate air, bone and ﬁre
keratinized hoof wall, so a soft tissue window of access is necessary to image deep

structures. Adair, et al, used laser Doppler ﬂowmetry to measure blood ﬂow at ﬁre

37
coronary band and dorsal hoof laminae in normal horses.” This technique is similar

to Doppler ultrasound, but it uses a beam of light in place of ﬁre sound wave. An 8
mm hole was cut in ﬁre dorsal hoof wall, to ﬁre junction of epidermal and dermal
laminae, to measure ﬂow in ﬁre underlying laminar vessels. This meﬁrod had
limited utility, as ﬁre window of access was limited to a small area and lirrrited in
duration of access due to drying of ﬁre laminae wiﬁr exposure to air. The effect of
mechanically drilling ﬁre hole may have altered local hemodynamics by
traumatizing ﬁre underlying tissues. This technique appears to be useful for
quantifying blood ﬂow at a restricted but specific location, while angiography
allows for visualization of vessels over a larger ﬁeld of view.

Magnetic resonance imaging (MRI) uses magnetic forces to cause vibration of
ﬁre atomic particles wiﬁrin ﬁre body tissues} 4‘ Each particle vibrates at a speciﬁc
frequency, characteristic of ﬁre atom and the magnetic ﬁeld. The contrast between
tissues is altered by varying ﬁre sequence of pulsation of one magnetic ﬁeld or by ﬁre
addition of gadolinium as an intravenous contrast material.“ In MRI angiography,
ﬁre movement of blood helps to differentiate ﬁre vessels from ﬁre surrounding
tissues.” Many artifacts are caused by turbulent or slow ﬂowing blood ﬁrat interfere
wiﬁr image interpretation and MRI angiography produces images of lesser
resolution ﬁran radiographic angiography so ﬁris newer modality is not considered
as ' 'c as contrast radiography for delineation of vascular lesions.”

Nuclear scintigraphy is also a non-invasive modality for imaging vascular

ﬂow patterns, ﬁrat records gamma radiation emitted from ﬁre patient, following

38
administration of a radioactive pharmaceutical.48 Nuclear angiocardiography

follows ﬁre vascular distribution of a radioactive material, 99mTechnetium, after a
peripheral venous injecﬁon. The energy emitted from ﬁre patient is detected by a
large crystal in ﬁre camera, creating a signal ﬁrat is transmitted to a dedicated
computer for image production. First pass nuclear angiocardiography (FPNA)
monitors activity during ﬁre initial phase of radioactivity ﬁrrough ﬁre heart.
Specialized computer programs are ﬁren used to calculate several cardiac function
irrdices, such as left ventricular ejection fraction, cardiac output, valvular regurgitant
fraction and pulmonary / systemic perfusion ratios.“ 49 This technique is effective for
identiﬁcation of cardiac abnormalities, such as, valvular insufﬁciencies, intracardiac
shunts and myocardial failure.49 Gaited equilibrium nuclear angiocardiography
(GENA) uses an ECG signal to trigger data acquisition at a selected phase of ﬁre
cardiac cycle, after ﬁre radiopharmaceutical has distributed ﬁrroughout ﬁre vascular
pool. This meﬁrod accumulates information over many cardiac cycles, increasing
ﬁre activity counts for each image. Patient motion is ﬁre limiting factor in ﬁris
procedure, since data are summated over a prolonged period of time.” Nuclear
angiography provides visual and quantitative information about ﬁre physiological
distribution of a radiopharmaceutical in ﬁre body by a relatively non-invasive
procedure as compared to cardiac caﬁreterization for selective radiographic
angiography. However, ﬁre anatomical detail produced in scintigraphic images is
not as well deﬁned as ﬁrose produced by digital radiography. Regions of tissue may

be evaluated for blood perfusion by scintigraphy, but individual vessels are not

39
distirrcﬁy delineated by ﬁris techniques0 In summary, there are advantages to each

of ﬁrese alternative imaging modalities, but not one of ﬁrem provides images of

small peripheral vessels ﬁrat are equal or superior in detail to contrast enhanced

digital angiography.

Angiography of the distal limb

The vascular supply of ﬁre distal limb has been investigated in several
species, using contrast radiography!”6 The normal angiographic appearances of
ﬁre bovine51 and caprine’2 digits have been reported. Gogoi, et al, examined ﬁre
blood supply to normal and abnormal bovine feet, in anesﬁreﬁzed patients.51 The
median artery (ﬁroracic limb) or dorsal metatarsal artery (pelvic limb) were
caﬁreterized, followed by intraarterial injection of 15-20 ml sodium ioﬁralamate, an
ionic contrast agent. Signiﬁcant alterations in ﬁre vascular pattern were observed in
various paﬁrological conditions. i

Angiography has been performed in normal and lamirritic horses, to assess
distal limb perfusion.5356 In all of ﬁrese studies, ﬁre paﬁents were anesﬁretized for
caﬁreterization of a distal limb artery and injection of a radiopaque solution.
Radiographs of ﬁre distal limb were produced at various time intervals following
contrast media injecﬁorr, wiﬁr ﬁre limb in various positions. Coffman, et al,
performed angiovenography on four horses, before and after production of
alimentary-induced acute lamirritisél3 Sodium and meglumine diatrizoate

(Hypaque®), 20-30 ml, was injected via ﬁre common digital artery for each sequence

40
of radiographs. Angiograms were performed in ﬁre dorsopalmar and mediolateral

views, using high speed ﬁlm and high speed intensifying screens. While ﬁris ﬁlm-
screen combination was useful to minimize paﬁent motion artifact and to decrease
radiation exposure, ﬁris technique was done at ﬁre expense of image dehi].1 In an
anesﬁretized patient, when moﬁon is not a signiﬁcant concern, sharper images
would. have been produced wiﬁr slower speed, dehil screens. In ﬁre normal feet,
each study demonstrated contrast enhancement of ﬁre medial and lateral digihl
arteries, terminal arch, several branches ﬁrrough ﬁre dish] phalanx, circumﬂex
artery and arteriovenous plexuses at ﬁre digital cushion and coronary corium.
Images of ﬁre laminitic feet demonstrated dilation of ﬁre arteries to ﬁre proximal and
middle phalanges and dilation of ﬁre digital arteries proximal to ﬁre terminal arch.
Therewasamarked decreaseinsiZeofﬁreterminalarteries and poorﬁlling ofﬁre
arteriovenous plexuses of ﬁre hoof. It was concluded ﬁrat ﬁre digihl arteries
constrict at ﬁre enhance of ﬁre volar foramina during acute lamirritis, ﬁrat blood
pools in capillary beds and ﬁrat blood is shunted ﬁrrough oﬁrer vessels of ﬁre dish]
limb.” Angiographic images from ﬁre same horses, before and after induction of
lamirritis, provided useful information for comparative purposes.

Clinically normal and chronically lamirritic feet have been examined by
angiography, using sodium and meglumine diatrizoate (Hypaque®) for vascular
contrast.54 Radiographs were hken at a consistent technique using par speed ﬁlm
and high speed screens. The arterial vessels visualized in ﬁre clinically normal

horses included ﬁre digihl arteries, ﬁre terminal arch and its branches to ﬁre dish]

41
margin of ﬁre ﬁrird phalanx and ﬁne vessels in ﬁre corium of ﬁre coronary band. In

ﬁris paper, Ackerman, et al, reported ﬁre diameter of ﬁre digihl arteries to be 0.4 cm,
hperingto02-0.3cmatﬁretermirralareh and ﬁrerewere 8-10 branches from ﬁre
munal arch ﬁrat were 0.1-0.2 cm in diameter. In ﬁre lamirritic subjects, filling of ﬁre
terminal arch and its branches was variable, wiﬁr some studies showing increased
size and decreased number of arterial branches while other studies had poor ﬁlling
of ﬁre terminal arch and normal branches along wiﬁr focal avascular regions. A
pattern of alternating vascular widening and narrowing was observed in one
lamirritic study, however histopaﬁrological examinaﬁon of ﬁris artery did not reveal
any abnormalities. Arterial caﬁreterization and vessel irrihtion from ﬁre contrast
material may cause vascular spasms, which may have accounted for ﬁris pattern.”
Ackerman, et a], concluded ﬁrat chronically lamirritic feet were characterized by an
irregularvascularpatterninﬁrecorium andbypoorﬁllingofﬁreterminalarchin
comparison to ﬁre symmetrical pattern observed in normal feet.“ While ﬁris
information is ﬁre only study ﬁrat describes ﬁre diameter of arteries in ﬁre equine
foot based on ﬁreir angiographic appearances, no information was provided
regarding ﬁre meﬁrod of measurement or ﬁre conﬁibuting factors of radiographic
magniﬁcation on vessel size. This information on measurement and radiographic
technique is imporhnt in order to compare and assess ﬁre results of ﬁris studies to
oﬁrer research or clinical dah.

In a study designed to examine ﬁre collateral blood ﬂow to ﬁre equine foot,

Scott, et al, ligated ﬁre medial palmar arteries and medial palmar digihl arteries of

42
seven ponies, ﬁren performed angiographic studies under general anesﬁresia.55

Meglumine ioﬁralamate was ﬁre contrast agent used to identify changes in ﬁre
vascular pattern and ﬁlling time associated wiﬁr arterial ligation. Collateral vessels
were identiﬁed on ﬁre angiograms ﬁrat mainhined blood ﬂow to ﬁre dish] limb
following ligation of ﬁre arteries. Immediately after ligation of ﬁre medial palmar
artery, ﬁre second and ﬁrird dorsal and palmar mehcarpal arteries were found to be
ﬁre main alternate channels of collateral blood ﬂow and ﬁre lateral palrrrar artery
continued to supply blood to ﬁre temrinal arch. This study demonstrated ﬁre ability
of ﬁre equine digit to compensate for vascular occlusion by using collateral vessels to
mainhin perfusion to ﬁre foot.55

In a clinical case report, Scott, et al, used angiography to assess vascular
perfusion of ﬁre dish] limb in ﬁrree horses. These angiograms demonstrated ﬁre
presence of vascular lesions, including partial to complete obstruction of ﬁre medial
and lateral palmar digihl arteries and rehined contrast media wiﬁrin ﬁre corium.
Several possible causes were proposed for ﬁrese observations, including decreased
arterial pressure, increased venous pressure, vasoactive factors, reduced frog and
digihl pressures and alterations in blood ﬂow due to heparirrizartiorré‘5

Van Kraayenburg, et al, performed peripheral angiographic studies in
conscious, shnding horses and dish] limb specimens to assess ﬁre effect of vertical
force on blood ﬂow in ﬁre palmar digihl arteries. This report demonstrated
interruption of blood ﬂow at ﬁre level of ﬁre dish] sesamoid bone and in ﬁre solar

canal under conditions of high vertical force. Cessation of blood ﬂow was absent in

43
limb specimens ﬁrat were not under stressed conditions. The intraarterial caﬁreter

was advanced beyond ﬁre bifurcation of ﬁre medial palmar artery to enter one
branch of ﬁre palmar digihl artery, so ﬁrat injected contrast medium entered ﬁre
terminal arch from only one direction. I-Iistologic sections of ﬁre palmar digihl
artery were hken at ﬁre level of ﬁre constriction and wall ﬁrickness was measured to
deternrine percenhge wall ﬁrickening and circumferencial luminal reduction. The
angiographic images were useful for subjective assessment of luminal patency, but
quantitative measurements were acquired only from ﬁre histologic secﬁons and not
from ﬁre radiographic images of inhct vessels in ﬁre live horses or limb specimens.58

All of ﬁrese reports were attempts to document ﬁre angiographic appearance
of normal and lamirritic feetl'wéﬁ8 They were performed using conventional
ﬂuorescent screens wiﬁr ﬁrick phosphor layers ﬁrat decreased image sharpness. In
contrast, ﬁre ﬂuorescent screens in digihl radiography, wiﬁrin ﬁre image intensiﬁer
tube, have a ﬁrinner phosphor layer wiﬁr more cryshls, designed to improve inrage
sharpness. The xray tube is constructed to provide a high amperage so ﬁre exposure
time may be kept short. Boﬁr of ﬁrese factors contribute to improved image dehil

and resolution of small vessels wiﬁrin and around ﬁre dish] phalanx of ﬁre horse.

STATEMENT OF HYPOTHESIS
Development of a technique for visualization of ﬁre intricate vessels of ﬁre

equine foot and measurement of ﬁreir diameters would be useful for evaluation of

44 .
vascular perfusion to ﬁre dish] limb. There are two hypoﬁreses eshblished for ﬁris

ﬁresis. First, ﬁre tohl magniﬁcation factor for a digihl radiographic imaging system
may be quantiﬁed by two meﬁrods, 1) comparison of a radiopaque object to
measurements of its radiographic image, and 2) calculation of individual
magniﬁcation factors for each component of ﬁre DR system based on physics of
radiography, ﬁren multiplication of ﬁrese factors to determine ﬁre tohl magniﬁcation
factor. The results of ﬁrese two meﬁrods for determining ﬁre tohl magniﬁcation
factor should be equal. The second hypoﬁresis shtes ﬁrat arterial vessels of an in-
vitro model of ﬁre equine foot may be idenﬁﬁed and vessel lurrrinal diameter may
be measured on angiographic images of ﬁre limb in two positions. Correction of
ﬁrese measurements for image magniﬁcation in each view should result in equal

values for luminal diameter of each vessel between ﬁre two positions.

METHODS AND MATERIALS

MAGNIFICATION FACTORS

Two meﬁrods were employed to calculate ﬁre tohl magnification factor of ﬁre
imaging system, used in ﬁre present study. The ﬁrst meﬁrod used a radiopaque
mehllic object of known size and ﬁre radiographic images of ﬁris marker to
determine tohl magniﬁcation (M) from ﬁre formula:

M 8 image size + object size.1

The second meﬁrod calculated each component of tohl magniﬁcation
indeperrdenﬁy, ﬁren combined ﬁrese factors to determine tohl magniﬁcation (M)
from ﬁre formula:

M=ngM.pr.

M - total magniﬁcation
M.- elecﬁ'orric magniﬁcation
Mp' plebmr’hic msrﬁﬁcation

The tohl magniﬁcation factors determined by ﬁrese two procedures were

compared to identify instrument settings under which ﬁrey produce similar results.

45

46

Meﬁrod 1

The widﬁr of a rechngular mehllic marker was measured wiﬁr digihl
calipers' and ﬁris value was called ﬁre object widﬁr. The marker was placed on ﬁre
patient hble of a General Electric ADVANTXG’ radiography unit" and imaged wiﬁr
ﬁre digihl radiography system, at 65 kVp, 100 mA, photoﬁmer controlled exposure
time and 0.3 mm focal spot. The intensiﬁer tube was positioned at 25, 30, 35, 38, 40
and 45 cm above ﬁre hble and ﬁre intensiﬁer tube was in ﬁre 9 inch, 6 inch and 4.5
inch modes of function (Figure 14). These images were printed at 1:1, 4:1, 6:1 and
12:1 formats by ﬁre laser imagerc to ﬁt 1, 4, 6 or 12 images, respectively, on each sheet
of 14x17" sheet of ﬁlm,d ﬁren ﬁre ﬁlm was developed in an automaﬁc processor:3
The widﬁr of ﬁre marker was measured on each radiograph ﬁrree times wiﬁr digihl

calipers and ﬁrese values were called ﬁre image widﬁrs.

 

" Digomatic calipers, Model M500=351, Mitutoyo, Japan

" General Electric SFX II ADVANTX", GE Medical Systems, Milwaukee, WI

° Laser Imager m, 3M Medical Imaging Systems Division, St. Paul, MN

" Infrascan" MC Laser Imaging film, DuPont Medical Systems, Wilmington, DE
' Kodak RP X-OMA'I", Model M68, Eastman Kodak Company, Rochester, NY

47

run L
DCALSPOT- —-—'-_‘—-

morsrm

 

 

 

gm::222:::::

Figure 14 Digihl radiographic images of a mehllic marker were produced with ﬁre marker on the
patient table.

The tohl magniﬁcation factor (M) for each focal-ﬁlm dishnce, intensiﬁer tube
mode and printer format were calculated, using ﬁre object widﬁr and ﬁre image
widﬁrs in ﬁre formula:

M=image widﬁr + object widﬁr
Since ﬁrese marker images incorporated all of ﬁre magniﬁcation effects of digihl
radiography, these values for magniﬁcation factors represented tohl magniﬁcation

of ﬁre system.

Meﬁrod 2
The ﬁrree components to image magrriﬁcaﬁon in digihl radiography include
geometric, electronic and photographic magnification. The magniﬁcation factor

from each component was calculated individually, based on ﬁre theories of image

48
magniﬁcation in radiography and photography, using dimensions of ﬁre imaging
system obhirred from ﬁre speciﬁcations manual for ﬁre General Electric ADVANTX"

imaging system’, GE service engineers12 or measured direcﬁy from ﬁre equipment.

Geometric magniﬁcation

The xray tube was positioned at a ﬁxed dishnce below ﬁre hble, so focal-
object dishnce (FOD) was 49.5 cm for every image. Focal-ﬁlm dishnce was ﬁre sum
of FOD and ﬁre dishnce from ﬁre hble to ﬁre intensiﬁer tube (25, 30, 35, 38, 40 and
45 cm). Using ﬁrese values for FOD (49.5 cm) and FFD (74.5, 79.5, 84.5, 87.5, 89.5,
94.5 cnr), geometric magnification (M3) was calculated for each position of ﬁre
intensiﬁer tube from ﬁre formula:

M¢=focal-ﬁlm dishnce + focal-object dishnce.1

Electronic magniﬁcation

Electronic magniﬁcation occurred wiﬁrin ﬁre image intensiﬁer tube, when ﬁre
electronic signal was focused on ﬁre output screen by ﬁre electroshtic focusing
lenses. The intensiﬁer tube input screen was 22.0 cm in diameter and ﬁre output
screen was 2.20 cm in diameter, so all images were magniﬁed by a factor of 0.10, or
miniﬁed by 10X, across ﬁre tube. The electrostaﬁc focusing lenses altered ﬁre focal
point of the electronic signal to cause additional magniﬁcation of ﬁre image which
varied wiﬁr ﬁre intensiﬁer tube mode of function. In ﬁre 9 inch mode, all of ﬁre

electrons from ﬁre input screen were focused onto ﬁre output screen, so ﬁre

49

magniﬁcation factor was 1.0. In ﬁre 6 inch mode, only ﬁre central 66% of ﬁre input
screen electrons were included on ﬁre output screen, so ﬁre image was magniﬁed by
a factor of 1.5 and in ﬁre 4.5 inch mode, ﬁre central 50% of ﬁre electron signal was
magniﬁed by 2.0x to ﬁll ﬁre output screen. Thus, electronic magniﬁcation (M.) in
ﬁre 9 inch mode was 0.10 (0.10 x 1). In ﬁre 6 inch mode Me was 0.15 (0.10 x 1.5) and

in ﬁre 4.5 inch mode Me was 0.20 (0.10 x 2).

Photographic magniﬁcation

Photographic magniﬁcation occurred from ﬁre output screen of ﬁre intensiﬁer
tubetoﬁrelaserimager,whereﬁrelaserbeamwasdirectedtoexposeﬁreﬁlm. The
output screen was 2.20 cm in diameter and ﬁre radiographic image diameter was
28.0 cm in ﬁre 1:1 format, 14.0 cm in ﬁre 4:1 format, 12.2 cm in ﬁre 6:1 format and 9.0
cm in ﬁre 12:1 format. Magniﬁcation is equal to image size divided by object size, so
photographic magniﬁcation (Mp) was determined by ﬁre formula:

Mp=radiographic image diamaer + output screen diameter.

Total magniﬁcation
Total magniﬁcation factors were calculated by ﬁre product of geometric,
electronic and photographic magniﬁcation factors:

M=M8 x M. x Mp,

50
Shtistical analysis

In Meﬁrod 1, ﬁre average of ﬁuee repeated measurements of each image was
used for image size, to determine ﬁre tohl magniﬁcation factors. These values for
ﬁre measured tohl magniﬁcation factors (TMFm) were analyzed by ﬁre linear
models procedure, multiple regression analysis.f Univariate and multivariate
analyses were performed on ﬁre independent variables, including focal-ﬁlm
dishnce (FFD), image intensiﬁer tube mode of function (II mode) and laser printer
format (format).

The calculated tohl magnification factors (TMFc) from Meﬁrod 2 were
compared to ﬁre TMFm values from Meﬁrod 1 by ﬁre paired T-test. A variable,
delh, was deﬁned as ﬁre difference between ﬁre magniﬁcation factors from ﬁrese
two meﬁrods (delh= Mm-TWC) for furﬁrer analysis. A linear models procedure,
multiple regression analysis of delh was performed to identify relationships
between ﬁrese independent variables and delh.

The shtistica] models built by ﬁrese regression analyses were used to
determine ﬁtted values for Mar and delh and ﬁreir residuals were tested by ﬁre

Wilk-Shapiro test for normal distribution.

 

fS'TA'I'IS‘I'IX 3.1, Analytical software, St. Paul, MN

51

TIME PERFUSION AND ANGIOGRAPHY
Irr-vitro perfusion model

The right ﬁroracic limb was removed, on each of six horses, at ﬁre mid-
diaphyseal region of ﬁre ﬁrird mehcarpus, inrmediately following euﬁranasia wiﬁr a
barbiturate overdose (Table 1). The media] palmar artery was caﬁreterized from ﬁre
severed end of ﬁre vessel, by advancing a piece of polyeﬁrylene tubing 5.0 cm dishd.
The caﬁreter was 10 cm long, wiﬁr an internal diameter 1.67 mm and an external
diameter 2.42 mm (PE 240), connected to a 16 g intravenous needle. The vessels
were ﬂushed wiﬁr 120 ml heparinized saline (2000 units/100 ml saline) and ﬁre

external surfaces of ﬁre limb were cleansed of debris.

Table 1 Following euﬁranasia, ﬁre right ﬁroracic limb was removed on each of six horses.

HORSE BREED AGE BODY WEIGHT GENDER
Grade 3 420
3 340
Arabian 4 370
4

400
Grade 365
4 320

 

Wiﬁrin ten minutes of euﬁranasia, ﬁre limb was placed in a shnding position
on ﬁre digihl radiography patient hble, supported by a ﬁrree prong clamp and ring
shnd. The support prongs conhcted ﬁre limb over its dorsolateral and dorsomedial
cuhneous surfaces, to avoid interference wiﬁr ﬁre palmar vessels. The

mehcarpophalangeal joint was in slight extension (approximately 10° from

52

perpendicular to ﬁre hble) to allow collection of venous efﬂuent, draining from ﬁre
mehcarpal veins. The collection bin was placed palmar to ﬁre foot to be excluded
from ﬁre imaging ﬁeld. The arterial caﬁreter was connected to a pump! by
polyeﬁrylene tubing (internal diameter 2.0 mm) and a short segment of an
intravenous ﬂuid line, including an injection port. The foot vasculature was
perfused wiﬁr a Krebs Henseleit solution (Appendix A), ﬁrat was maintained at 37°C
and aerated wiﬁr 95% 02 and 5% C02 (Figure 15). This isolated limb perfusion
technique was adapted from ﬁre meﬁrod reported by Elmes and Eyre for
investigation of vascular reactivity in ﬁre bovine foot.59 The pump ﬂow rate was
irriﬁally set to perfuse ﬁre limb at a ﬂow rate of 150 ml/ min. This rate was calibrated
at position 8 on ﬁre pump setting dial and measured by ﬁre volume of solution
pumped into a graduated cylinder in one minute. A pressure monitorh was
connected between ﬁre pump and foot to register arterial perfusion pressure. The
pump ﬂow setting was adjusted between 150-175 ml/ min. to mainhin a perfusion
pressure of 100 mm Hg.60 The limb was perfused for ﬁve rrrinutes at ﬁris ﬂow rate to

demonstrate a steady shte of vessel diameter as determined by a consistent arterial

perfusion pressure of 100 mm Hg.

 

9 Masterﬂex", Mode] 7014, Cole Palmar
" Simultrace Recorder, VR 12/ 16, Honeywell, Pleasantville, NY

53

37‘C injection port

   
   
   
 

catheter
in medial
palmara.

 

 

 

 

 

100mm Hg

Jeans newsman Pump

L SOLUTION I

 

 

 

 

 

 

 

Pressure
Heater Monitor

 

 

 

 

 

 

Figure 15 The vaculature was perfused wiﬁr an oxygenated (95% O2/5%C02) , heated (370C) Krebs
Hemeleit solution at 1(1) mm Hg arterial perfusion pressure.

Angiography

Angiographic studies were performed on each limb, in ﬁre shnding and
lateral recumbent posiﬁons. In boﬁr studies, ﬁre radiographic technique was 65 kVp,
100 mA, phototimer controlled exposure time and 0.3 mm focal spot. The contrast
medium was 10 ml of iopamidol 6196‘ (37°C), injected as a rapid bolus, via ﬁre
injection port in ﬁre tubing. Images were acquired at one frame per second for 25
seconds, beginning at ﬁre initiation of ﬁre contrast medium injecﬁon. Changes in
arterial perfusion pressure following injection of ﬁre contrast medium were noted.
All of ﬁre palmarodorsal studies and ﬁve (5/ 6) of ﬁre lateromedial studies were
recorded in ﬁre 9 inch mode of ﬁre image intensiﬁer tube. One lateromedial study

(horse 1) was recorded in ﬁre 6 inch mode. All of ﬁre images were printed at a

 

1 Isovue-300', 300 mg 1/ ml, Squibb Diagnostics, Princeton, NJ

54
consistent window widﬁr (255) for image contrast and window level (110-125) for

image brightness in ﬁre 6:1 format and developed in an automatic processor.

Palmarodorsal view

The ﬁrst angiogram was performed wiﬁr ﬁre limb in a shnding position, on
ﬁre hble This was called ﬁre palmarodorsal (PD) view due to ﬁre direction of xray
transmission ﬁrrough ﬁre foot, from ﬁre xray tube below ﬁre hble, to ﬁre image
intensiﬁer tube above ﬁre foot (Figure 16). The image intensiﬁer tube was
positioned 38.0 cm above ﬁre hble, wiﬁr ﬁre foot centered in ﬁre ﬁeld of view. This
hble to intensiﬁer tube dishnce was selected to avoid interference of ﬁre limb
preparation by ﬁre intensiﬁer tube. The xray tube was located in a ﬁxed position,

49.5 cnr below ﬁre hble, so focal-film distance for ﬁre PD study was 87.5 cm.

 

 

 

 

 

38.0:-

 

 

 

 

‘7'5 a as 0: l I I

49.5 arr

 

 

 

Figure 16 The palmarodorsal study was performed wiﬁr ﬁre limb in a shnding position , FFD=87.5
cm, FOD-49.5 an and object-ﬁlm dishnce=38.0 cnr.

55
lateromedial view
Following ﬁre palmarodorsal study, each limb was repositioned in lateral
recumberrcy for ﬁre lateromedial (LM) study. The arterial perfusion pressure
decreased during repositioning so ﬁre perfusion rate was increased to approximately
175 ml/min to mainhin arterial perfusion pressure at 100 mm Hg. The image
intensiﬁer tube was positioned 25.0 cm above ﬁre hble and ﬁre xray tube remained

at 49.5 cnr below ﬁre hble, resulting in ﬁre new FFD of 74.5 cm (Figure 17).

rag,

 

 

 

 

 

 

 

 

785 car a. 0' l l
. a

49.5:-

 

 

 

Figure17 Thehteromedialstudywasperformedwiﬁrﬁrelimbinlateralrecumbency, FFD=745<m,
FOD-49.5 cm and object-ﬁlm dishnoe=25.0 an.

Vessel identiﬁcation and measurement
Vessel identiﬁcation
Anatomy of ﬁre normal blood vessels of ﬁre equine foot was examined, using .

several published referencesﬁu” Four equine ﬁroracic dish] limb specimens were

56

perfused wiﬁr Batson's solutioni followed by soft tissue digestion by a colony of
Dermestm lardius and Dermatw vulpinus beeﬁes to create corrosion casts of ﬁre
digihl arteries. Two additional limbs were perfused wiﬁr red colored latex and
permitted to solidify for ﬁrree days prior to dissection to visualize ﬁre arteries. These
limb preparations were from ﬁre same horses as ﬁrose used in ﬁre perfusion studies
and served only to familiarize ﬁre auﬁror wiﬁr ﬁre vascular anatomy. No attempt
was made to measure and correlate ﬁre size of ﬁre plastic column wiﬁrin ﬁre
dissected vessels wiﬁr ﬁre vessels observed radiographically. Table 2 is a list of ﬁre
arterial vessels identiﬁed on angiograms in ﬁre palmarodorsal and lateromedial

studies.

Table2. 'I'hepalmardigitalarteryanditsmajorbrarrchestothefootwereidentiﬁedondigital
angiograms.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Common rrame‘1 Nomina Anatomica Veterinaria Terminolm‘z
palmar digital a. A. digitalis palmaris communis II
bulbar a. Ramus tori digitalis
coronary a. A. coronalis
dorsal phalangeal a. Ramus dorsalis phalangis proximalis
palmar phalangeal a. Ramus palmar-is phalangis proximalis
distal dorsal phalangeal a. Ramus dorsalis phalangis distalis
distal palrrrar phalangeal a. Remus palmaris phalangis distalr's
dorsal branch of distal dorsal phalangeal a. Remus dorsalis of ramus dorsalis phalangis distalis
pahrrar branch of dish] dorsal phalangeal a. Ramus palmar-is oframus dorsalis phalangis distalis
terminal arch Arcus terminalis
dorsalbranchssofterminalarch Ranmsdorsalisaraisterminalis
solarbrandresofterminalarch Ranmspalmarisarcusterminals
solar margin a. Ramus marginis solarris

 

 

j Batson's No. 17 Plasﬁc Replica and Corrosion Kit, Polysciences, Inc., Warrington,
PA

 

58

Vessel measurement
The tohl magniﬁcation factors determined by comparison of a mehllic

marker widﬁr to its radiographic image widﬁr (meﬁrod 1) were used to calculate
vessel diameters, since boﬁr ﬁre marker images and the vessel images were
measured wiﬁr ﬁre digihl calipers. The formula:

Magniﬁcation factor = image size + object size
was rewritten to solve ﬁre actuation for object size:

Object size = image size + magniﬁcation factor
Using ﬁris formula, ﬁre sizes of ﬁre vessels wiﬁrin ﬁre body were determined by ﬁreir
radiographic images. Vessel size measurements from ﬁre radiographs were divided
by ﬁre tohl magniﬁcation factor, for ﬁre appropriate FFD, intensiﬁer tube mode and

printer format to determine ﬁre actual vessel widﬁrs.

Shtistical analysis

The repeahbility of ﬁris angiographic technique was described by ﬁre
frequency of vessel identiﬁcaﬁon as a percenhge of ﬁre six feet examined.
Descriptive shtistics of ﬁre vessels included mean values for each vessel diameter, in
each position, averaged for ﬁre six feet and reported wiﬁr ﬁre shndard deviation of
ﬁre mean. Measurements of vessels ﬁrat were only identiﬁed on one view, were
discarded for furﬁrer shtistica] analysis. The vessel diameters, as determined from
ﬁre digihl angiographic images in ﬁre palmarodorsal and lateromedial views were

compared by ﬁre paired T-test (P< .05).

57

The diameter of ﬁre contrast enhanced lumen of ﬁrese vessels were measured
with digihl calipers on one image of each foot in ﬁre palmarodorsal and
lateromedial angiograms. The images were selected from ﬁre late arterial perfusion
phase prior to contrast enhancement of ﬁre small capillaries and venules of ﬁre foot.
All ﬁre measurements were performed wiﬁrin a 24 hour period to mainhin
consistency in ﬁre technique. In boﬁr views, ﬁre palmar digihl artery was measured
at ﬁrree locations: 1) 1.0 cm proximal to ﬁre bulbar artery, 2) 1.0 cm proximal to ﬁre
dorsal phalangeal artery and 3) at ﬁre entrance of ﬁre solar canal of ﬁre ﬁrird phalanx
(Figure 18). The terminal arch and solar margin arteries were measured at ﬁre
sagithl rrridline and ﬁre remaining vessels were measured 1.0 cm dish] to ﬁreir

origin from ﬁre palmar digihl artery.

 

 

 

 

Pigure18 Thepalmardigitalarterywasmeasuredatﬁrreelocaﬁonszl)1.0cmproximaltotheorigin
ofﬁrebulbara.,2)1.0cmprordmaltoﬁreoriginofﬁredorsalphalangeala.and3) atﬁreentranceto
ﬁresolarcanal.

RESULTS

MAGNIFICATION FACTORS

Toh] magniﬁcation factors (TMFm) determined by comparison of the size
of a mehllic object to measurements of its radiographic image, are presented in
Table 3, for each combination of focal-ﬁlm dishnce (FFD), image intensiﬁer tube
mode (II mode) and printer format (format). The image widths from the three
repeated measurements of each radiographic image wiﬁr the mean and shndard
deviation of the mean values are listed in Appendix B. Image magniﬁcation
factors ranged from 0.70 (at ﬁre shortest FFD, largest II mode and highest
format) to 5.41 (at ﬁre longest FFD, smallest II mode and lowest format).

Several trends were identiﬁed in ﬁrese magniﬁcation factors, which
related to the independent variables, FFD, II mode and printer format. As FFD
increased, the magniﬁcation factors increased slighﬁy (Figure 19). As the II
mode increased, ﬁre tohl magniﬁcation factors decreased markedly (Figure 20)
and as ﬁre number of images printed per sheet of ﬁlm (format) increased, ﬁre

tohl magnification factor decreased (Figure 21).

59

60

Table 3 Total magniﬁcation factors (TMFm) were determined by measurement of a metallic
marker and its radiographic image.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

II MODE 9 inch PRINTER FORMAT
FFD (cm) 1:1 4:1 6:1 12:1
74.5 2.15 1.07 0.93 0.70
79.5 2.27 1.13 0.97 0.73
84.5 2.39 1.18 1.03 0.77
87.5 2.47 1.22 1.05 0.79
89.5 2.54 1.25 1.09 0.81
94.5 2.65 1.31 1.14 0.86
II MODE 6 inch PRINTER FORMAT
FFD (cm) 1:1 4:1 6:1 12:1
74.5 3.10 1.54 1.33 1.01
79.5 3.28 1.62 1.39 1.08
84.5 3.47 1.72 1.50 1.12
87.5 3.61 1.77 1.54 1.15
89.5 3.69 1.82 1.59 1.18
94.5 3.88 1.90 1.65 1.24
11 MODE 4.5 inch PRINTER FORMAT
FFD (cm) 1:1 4:1 6:1 12:1
74.5 4.35 2.14 1.85 1.40
79.5 4.57 2.27 1.93 1.50
84.5 4.88 2.40 2.09 1.55
87.5 5.04 2.47 2.16 1.61
89.5 5.17 2.53 2.20 1.66
94.5 5.41 2.64 2.29 1.71

 

 

 

 

 

 

 

 

61

 

 

 

 

2.5
W
2 +
£1.5-
1..
0.5-
o I I I I I I I I I I I I I I I I I I ﬁj
“2 “'2 “2 “'2 “2 “2 “2 “2 “2
a :2 :2 s 3 s s a s 3 s
focal-filmdistance(cm)

Figure 19 The total magniﬁcation factors (TMFm) increased slightly as FFD increased.

 

 

 

3
25 }\
2 - o
§L5~ __,
1 ..
0.5 -
0 . T . u u T r r r
4.5 5 5.5 6 6.5 7 7.5 8 8.5 9
image intensifier mode

Figure 20 As 1] mode increased, the total magniﬁcation factors (TMFm) decreased steadily.

4
3.5
34
5125‘

2..
E15«
1. *0
05- '

o I I I I I I I I I ﬂ 1

1:01 2:01 3:01 4:01 5:01 6:01 7:01 8:01 9:01 10:01 11:01 12:01
‘ printer format

 

 

 

Figure 21 The total magniﬁcation factors (TMFm) decreased as the number of images per sheet
of ﬁlm (format) increased.

62

Statisﬁcal analysis determined that ﬁre tohl magniﬁcation factors (T MFm)
were signiﬁcanﬁy different at each FFD, II mode and format at P<.0001. Oﬁrer
signiﬁcant predictor variables included the interaction of FFD and format,
(format)2, (II mode)2 and (format)3. The formula for the linear function of
TMFm, as deﬁned by the linear models procedure, was:

Fitted TMFm = 8.968 + 0.021663(FFD) - 1.5701(FORMAT) - 1.4139(11 MODE) +

2.8479(FFD x II MODE) + 0.18996(FORMAT)2 + 0.069753(II MODE)2
- 0.0083247(FORMAT)3
The R2 value for this model was 97 .5% and ﬁre residuals were normally
distributed, as determined by the Wilk-Shapiro test (P=0.9472). The output of
each univariate and multivariate analysis is presented in Appendix B. The
values for FFD, II mode and format, used in ﬁris study, were entered into ﬁris
formula to calculate ﬁtted values for TMFm, based on the linear function.
Graphs of ﬁre ﬁtted TMFm values for each FFD, H mode and format are
presented in Appendix B. These graphs are almost identical to ﬁre experimenh]
data because the Shtistical model accounted for 97.5% of ﬁre variability in
TMFm.

Tohl magniﬁcation factors (TMFc) calculated for each FFD, II mode and
printer format, using formulas derived from ﬁre physics of radiography and .
photography, are listed in Table 4. The contributing magniﬁcation factors for
each component, geometric, electronic and photographic magniﬁcation, are

listed in Appendix C. TMFc was signiﬁcanﬁy smaller ﬁran TMFm for every

combination of FFD, II mode and format.

63

Table 4 Total magniﬁcation factors (TMFc) were calculated for the inraging system.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

II MODE 9 inch PRINTER FORMAT
FFD (cm) 1:1 4:1 6:1 12:1
74.5 1.92 0.96 0.83 0.62
79.5 2.05 1.02 0.88 ‘ 0.66
84.5 2.18 1.09 0.94 0.70
87.5 2.27 1.13 0.97 0.73
89.5 2.30 1.15 0.99 0.74
94.5 2.43 1.22 1.05 0.78
11 MODE 6 inch PRINTER FORMAT
FFD (cm) 1:1 4:1 6:1 12:1
74.5 2.88 1.44 1.24 0.92
79.5 3.07 1.54 1.32 0.98
84.5 - 3.26 1.63 1.40 1.05
87.5 3.40 1.70 1.46 1.09
89.5 3.46 1.73 ~ 1.49 1.11
94.5 3.65 1.82 1.57 1.17
II MODE 4.5 inch PRINTER FORMAT
FFD (cm) 1:1 4:1 6:1 12:1
74.5 3.84 1.92 1.65 1.23
79.5 4.10 2.05 1.76 1.31
84.5 4.35 2.18 1.87 1.39
87.5 4.53 2.27 1.95 1.45
89.5 4.61 2.30 1.98 1.48
94.5 4.86 2.43 2.10 1.56

 

 

 

 

 

There was no signiﬁcant difference in delh (TMFm-TMFc) over the six
focal-film dishnces (Figure 22). As ﬁre II mode increased, delh decreased

(Figure 23) and as printer format increased, delh decreased (Figure 24).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.5
2
1.5“ _lmFm
E UTMFC
1 mm
0.5
o ."'.”"”.'; """"

 

«2 v: we

 

focal-film dishnce (cm)

Figure 22 As FFD increased, the difference between TMFm and MC (delta) remained
unchangd.

 

 

 

 

 

 

 

 

 

 

 

 

4.5 5 5.5 6 6.5 7 7.5 8 8.5 9
image intensifier tube mode (inch)

Figure 23 As II mode increased, the difference between TMFm and MC (delta) decreased.

 

 

 

 

 

 

 

 

 

 

 

 

4
3 .1 » .
E:-
1 .
o " ‘“ I
1:01 2:01 3:01 4:01 5:01 6:01 7:01 8:01 9:01 10:01 11:01 12:01
printer format

Figure 24 As printer format increased, the difference between TMFm and TMFc (delta)
decreased.

65

Delh was signiﬁcanﬁy different for magniﬁcation factors at various II
modes and printer formats. Oﬁrer predictor variables included ﬁre interaction of
II mode and format, (II mode)2, (format)2 and (format)3. The formula for the
linear function of delta, as defined by ﬁre linear models procedure, was:

Fitted delh = 1.8004 - 0.37619(II MODE) - 0.17657(FORMAT)
+ 0.0029212(II MODE x FORMAT) + 0.024043(II MODE)2
+ 0.023579(FORMAT)2- 0.0010728(FORMAT)3

The R2 value for ﬁris model was 91.4% and the residuals were normally
distributed, according to ﬁre Wilk-Shapiro test (P=0.9720). The output of each
univariate and multivariate analysis is presented in Appendix C. The values for
FFD, II mode and format used in ﬁris study were entered into ﬁris formula to
calculate ﬁtted values for delta, based on ﬁris linear function. Graphs of ﬁre
ﬁtted delh values for each FFD, II mode and format are presented in Appendix

C. These graphs are quite similar to the experimenhl dah because the statistical

model accounted for 91.4% of ﬁre variability of delh.

66

ANGIOGRAPHY OF EQUINE DISTAL LIMB

The digihl radiographic technique produced adequate quality images for
consistent identiﬁcation of the palmar digital artery and its major branches to the
foot in the palmarodorsal and lateromedial views (Figures 25A and 26A). The
bulbar and coronary arteries were not included in the ﬁeld of view on most of
ﬁre PD images and the palmar digihl artery at the entrance of the solar canal
was not identiﬁed on the LM view in two feet, due to the poor image contrast
and overall image darkness. Oﬁrerwise, the percenhge of identiﬁcation was 83-
100% for the remaining arterial vessels (Table 5). Aﬁases of these vessels were

drawn from the angiographic images (Figures 25B and 26B).

Table 5 Arterial vessels were identiﬁed on angiograms of six equine feet.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VESSEL PD" LM‘
palmar digital a. proximal to bulbar a. 100% 100%
proximal to dorsal phalangeal a. 100% 100%
entering solar canal 100% 67% (4/6L
bulbar a. 0% (0/ 6) 100%
coronary a. 17% (1/ 6) 100%
dorsal phalangeal a. 100% 100%
almar phalangeal a. 83% (5/ 6) 100%
distal dorsal phalangeal a. 100% 100%
distal palmar phalangeal a. 100% 100%
dorsal branch of distal dorsal phalangeal a. 100% 100%
palmar branch of distal dorsal phalangeal a. 100% 100%
terminal arch 100% 83% (5/ 6)
dorsal branches of terminal arch 100% 100%
solar branches of terminal arch 100% 100%
solar margin a. 100% 83% (5/ 6)

 

 

 

1 Units are % of feet in which ﬁrese vessels were identiﬁed.

67

 

 

(A)

mam

A palmardlgltala

B bulbara.

C a.

D dorsalphalarrgeala.

E palmar a.

F distaldonalphalangeala.

G distalpalmarphalarrgeala.

H doualbnndrof distal dorsal phalangeal a.
l palnrarbrandrofdbta] dorsal phalangeal a.
I terminalardr

K donalbrarrdrofterminalardr

L solarbrandroftermlnalardr

M eolarmarglna.

Figure 25 The palmar digital artery and its major arterial branches to the equine foot were
visible on digihl angiograms in the palmarodorsal view (A) and diagrammed as an atlas of
these vessels (B).

68

 

 

 

(A) (B)
team

A palrnardlgitala.

B bulbara.

C coronarya.

D dorsalphalangeala.

E phalangeala.

F distaldoraalphalangeala.

G palmarphalarrgeala.

H dondbnndrofdtsta] dorsal phalangeal a.
l palmarbrandrofdistaldorsalphalangea] a.
I terminalardr

K dorsalbranclrofterminalardr

L aolarbrandroftermlnalardr

M aolarrrrargtna.

Figure 26 The palmar digital artery and its major arterial branches to ﬁre equine foot were
visible on digital angiograms in the lateromedial view (A) and diagrammed as an atlas of these
vessels (B).

69
A vascular bed was visualized at the dorsal aspect of the coronary region,
supplied by vessels from two directions. Dorsal branches of the terminal arch
ﬁlled with contrast material in the distal to proximal direction and distal
branches of ﬁre dorsal phalangeal artery enhanced from the proximal to distal
direction (Figure 27). This ﬁnding suggests an alternative route of vascular
supply to ﬁre dorsal aspect of the hoof via ﬁre dorsal phalngeal artery in case of a

decrease in perfusion from the terminal arch.

 

Figure 27 Dorsal branches of the terminal arch and distal branches of the dorsal phalangeal
artery supply a vascular bed at the dorsal aspect of the coronary region.

70
Some minor variations were observed in the pattern of ﬁre terminal arch

and solar branches, but ﬁre overall distribution of vessels was ﬁre same among

ﬁre six feet in ﬁris study (Figure 28).

as it as?

Horse 1 i Horse Horse 3

Figure 28 Minor variations were present in ﬁre pattern of ﬁre terrrrinal arch and solar branches,
as seen on ﬁre palmarodorsal view.

Alterations in arterial perfusion pressure, following injection of ﬁre
contrast medium, occured in two feet. In one foot, ﬁre presure rose rapidly from
100 to 350 mm Hg ﬁren gradually returned to its original pressure wiﬁrin ﬁrree
minutes. In a different foot, the opposite response occured with a rapid decline
in pressure from 100 to 70 mm Hg, followed by a gradual return to the initial
pressure over 3-5 minutes. In both feet, the same vascular reactions reoccurred

following the second contrast medium injection in lateral recumbency.

Vessel measurement

The mean size and standard deviation of the mean for each arterial vessel,
averaged over the six feet, are listed in Table 6. These values were corrected for
image magniﬁcation using ﬁre factors determined by method 1. For all of ﬁre PD
studies, the magniﬁcation factor was 1.05. In ﬁre LM studies ﬁrat were imaged in
the 9 inch II mode (horses 2-6), the magniﬁcation factor was 0.93 and in horse 1,
images were acquired in the 6 inch H mode, so the magniﬁcation factor was 1.33.
Image measurements and vessel sizes for each foot are listed in Appendix D. All
of ﬁre vessels were signiﬁcanﬁy larger on the lateromedial views, as determined
by the paired T-test (P<.05) except for ﬁre terminal arch, in which ﬁre

lateromedial view vessel sizes were signiﬁcanﬁy smaller (Figure 29).

Table 6 Arterial vessels were measured on digital angiograms in the palmarodorsal (PD) and
lateromedial (LM) views and vessel size was corrected for magniﬁcation.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VESSEL PD WEI. SIZE (mm) LM VF$EL SIZE (mm)
X :1: SD X :i: SD

palmar digital a. proximal to bulbar a. 2.28 :i: 0.55 2.54 :i: 0.48
Pmadmal to dorsal phalangeal a. 1.82 a: 0.43 2.14 :r 0.22

entering solar canal 1.61 a: 0.53 1.37 :i: 0.23

bulbar a. not available 1.53 :t 0.62
coronary a. 0.80 1.09 :t 0.35
dorsal phalangeal a. 1.14 :t 0.20 1.34 :i: 0.42
palmar phalangeal a. 0.73 :t 0.14 0.97 r: 0.20
distal dorsalphalangeal a. 1.11 i 0.30 1.71 :i: 0.70
distal palmar phalangeal a. 0.97 :i: 0.23 1.38 i: 0.82
dorsal branch of distal dorsal phalangeal a. 0.86 :i: 0.17 1.21 :i: 0.48
ar branch of distal dorsal phalangeal a. 0.75 :i: 0.15 0.90 :i: 0.31
terminal arch 1.42 :i: 0.24 1.80 :i: 0.51
dorsal branch of terminal arch 0.71 i: 0.17 0.87 :i: 0.35
solar branch of terminal arch 0.85 :t 0.23 0.87 i 0.33
solar margin a. 0.80 i 0.15 0.74 :i: 0.09

 

 

 

72

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

3
2.5-
5 2'
5 1.5- ,_ I- l-
'5
i
a 1- —
0.5d — *-
ou -.—- -.
A'A"A”'BCDEFGHIIKLM
vessel
Ingram E Weirdness!”
F distaldoraalphalarrgeala.
A palmardtgttala. G distalpalmarphalangala.
A' proximaltobulbara. H dorsalbranchofdishldorealphalangeala.
A" proximaltodoraalphalangeala l palmarbranchofdistaldoraalphalangeala.
A"'enteringsolarcanal I termirralarch
B bulbara. K dorsalbrarrdrofterminalarch
C a. I. solarbranchofterminalarch
D dorsalphalangeala M solarmargina.

Figure 29 Arterial vessels of the equine foot were measured on digital angiograms of six feet
and corrected for image magniﬁcation in the palmarodorsal (PD) and lateromedial (LM) views.

DISCUSSION

MAGNIFICATION

Magniﬁcation is inherent in radiography, yet there is not one shndard
meﬁrod for quantifying the magniﬁcation factor of an imaging system.
Comparison of a radiopaque marker to its image has been used as a crude
assessment of image magniﬁcation. However, critical examination of a
radiograph reveals an indistinct margin to the image that may cause an error in
measurement. This ill-deﬁned margin is called penumbra and it is due to the
various origins of xrays at ﬁre focal spot. Use of the smallest available focal spot
produces the highest quality image by minimizing the penumbra.1

A second method for quantifying the magniﬁcation of a radiographic
system was described in ﬁris ﬁresis, for comparison to ﬁre previous method.
Geometric magniﬁcation is commonly acknowledged in radiology and
quantiﬁed by the dishnces between the focal spot, object and ﬁlm. Penumbra
also contributes to geometric magniﬁcation and its magnitude may be calculated

from the formula:

Penumbra = focal spot size x object-ﬁlm dishnce + focal spot-object distance1

73

74

Due to the indistinct appearance of the margins of ﬁris penumbra zone, it is
difﬁcult to consistenﬁy include the errﬁre penumbra in the measurement of an
image (Figure 30). Therefore, ﬁris formula overestimates ﬁre magnitude of

penumbra ﬁrat is included in a measurement.

 

0'98

 

 

 

Figure 30 Penumbra contributed to image magniﬁcation, but due to the ill-deﬁned, gradually
fading margins, it was not completely included in the measurements.

The size of a focal spot is reported in the speciﬁcations manual of
radiographic equipment. The actual size of a focal spot will vary from its
reported size, depending on the radiographic technique.1 The National Electric
Manufacturers Association has eshblished shndards for techniques for

measurement of focal spot size and tolerance limits for deviation from the stated

75
size.1 The magnitude of penumbra also depends on the relative size of ﬁre
object and ﬁre focal spot. When an object is much larger ﬁran the focal spot,
penumbra is small (Figure 31A). When ﬁre object is smaller than the focal spot,

penumbra is larger than the umbra, rendering ﬁre image non-representative of

ﬁre original object (Figure 31B).

 

(A) ' ‘ (3)

Figure 31 The magnitude of penumbra depends on the relative size of ﬁre object and the focal
spot. Penumbra (P) is small when the object is larger ﬁran ﬁre focal spot (A) and penumbra is
largerﬁrarrﬁreumbra(Lthentlreobjectissmallerthanﬁrefooalspotm).

The quantihtive contribution of penumbra to tohl image magniﬁcation
was not included in ﬁre formula for ﬁre calculated tohl magniﬁcation factor
(MC) in meﬁrod 2, due to ﬁrese anﬁcipated complications. By intentionally
excluding penumbra from the equation, [TMFc values were consistenﬁy smaller

than ﬁre measured tohl magniﬁcation factors (T MFm). Had the theoretical

76

value for penumbra been included in the TMFc procedure, ﬁre TMFc values
would have been consistenﬁy larger than the TMFm values.

Penumbra had a deleterious effect on image quality that was observed
during measurement of the radiographic images of ﬁre mehllic marker. The
radiographic measurements of the mehllic marker, from ﬁre three repetitions in
method 1, indicated a slighﬁy greater standard deviation of the mean at the 4.5
inch II mode, as compared to the lesser magniﬁed images (Appendix B).

Electronic and photographic magniﬁcation signiﬁcanﬁy effected the ﬁnal
image size and their contributions were included in the calculaﬁon of TMFc.
These components are not involved in conventional radiography where the ﬁlm
is direcﬁy exposed in ﬁre cassette. However, in digihl radiography, ﬁre latent
xray image is converted to an electrical signal and manipulated by the system to
expose ﬁlm in a remote printer. Linear regression analysis deﬁned the
relationship between TMFm and geometric magniﬁcation (FFD), electronic
magniﬁcation (II mode) and photographic magniﬁcation (format). The
shtistically signiﬁcant predictor variables included, FFD, II mode, format, the
interaction of II mode and format, (II mode)2, (format)2 and (format)3. This
analysis demonstrated ﬁre importance of considering electronic and
photographic magniﬁcation in quantifying image magniﬁcation in digihl
radiography. Penumbra was part of ﬁre image at the input screen of ﬁre image

intensiﬁer tube. Therefore, alterations in image size, by electronic or

77
photographic magniﬁcation, also effected the size of penumbra on ﬁre ﬁnal

image.

There were distinctive trends in the disparity between TMFm and TMFc
values (delh) that related to FFD, II mode and format. While FFD was
signiﬁcant in predicting variability in TMFm, it was not was not signiﬁcant in
predicting variability in delh. The magniﬁcation factors only increased slighﬁy
over a wide range of focal-ﬁlm distances, so FFD had only a minute effect on
penumbra. II mode, format, their interaction (11 mode x format), (II mode)2,
(format)2 and (format)3 were signiﬁcant predictors of variability for both TMFm
and delh. The similarity between ﬁrese analyses suggests that a close
relationship existed between the magnitudes of image magniﬁcation and delh.
Penumbra was ﬁre link between ﬁrese factors. As the images were increasingly
magniﬁed, the penumbra was enlarged and ﬁre underestimation of TMFc
relative to TMFm increased.

Thus, if it is the intention to measure a radiographic image for
determination of tobject size, then a radiographic technique should be used that
minimizes penumbra in order to produce an image with sharp margins.
Penumbra is smallest when image magniﬁcation is at its least. In digihl
radiography, ﬁris may be accomplished by using a short focal-film dishnce
(assuming a ﬁxed focal-object dishnce), using the largest image intensiﬁer mode
and printing the greatest number of images on each sheet of ﬁlm. It must also be

recognized ﬁrat magniﬁcation may enhance one's ability to visualize small

78

structures, therefore, some compromise in technique is necessary depending on

ﬁre speciﬁc application.

ANGIOGRAPHY OF THE EQUINE DISTAL LIMB

The pattern of the arterial vessels observed in ﬁris study was consistent
with the described anatomy in the 1iterature.~"“”v"’2v‘54 Some minor variations were
present in ﬁre shape of the terminal arch and number of solar branches, but ﬁre
overall distribution of ﬁre vessels was the same among ﬁre six feet in ﬁris study.

It was possible to identify most of the arterial vessels in the palmarodorsal
and lateromedial views. In the PD view, the bulbar and coronary arteries were
usually outside the ﬁeld of view, while these vessels were included in the LM
images. The longer FFD used in the PD studies caused image magniﬁcation that
eliminated the peripheral vessels from the images. A decrease in the ﬁeld of
view is a disadvanhge of magniﬁcation ﬁrat must be considered when geometric
or electronic magniﬁcaﬁon techniques are used. However, photographic
magniﬁcation allows for the enlargement of the entire image by printing ﬁre
same information at larger dimensions. This technique is a unique feature of
digihl radiography.

The palmar phalangeal artery was not identiﬁed in one foot, due to its
parallel orienhtion relative to the xray beam. Imaging of ﬁre foot from multiple

projections was helpful for visualization of ﬁris small vessel and necessary for

79
localization of structures in a three-dimensional perspective. Repositioning the

limb preparation, from ﬁre shnding to ﬁre lateral recumbent position, caused
alteraﬁons in arterial perfusion pressure ﬁrat may have had a signiﬁcant effect
on the vessel diameter measurements. A radiographic technique for imaging of
an object from multiple projections, wiﬁrout repositioning the object would be
helpful to avoid ﬁris source of error. A C-arm unit« is a ﬂexible version of digihl
radiographic equipment that may be used to image an object from several angles
by rohting the xray tube and the image intensiﬁer tube without repositioning
the subject. Use of a C-arm unit wiﬁr ﬁris in-vitro foot model would eliminate
the changes in perfusion pressure ﬁrat occurred during reposiﬁoning of the limb.

The terminal arch and solar branches were easily identiﬁed on every PD
image but ﬁre dehil was less distinctive at ﬁre dorsal branches of terminal arch.
These dorsal branches were ﬁre smallest vessels measured in ﬁris study (mean
diameter = 0.71 :i: 0.17 mm) which was near the limit of resolution for a 0.3 mm
focal spot.1 Thus, despite the use of the smallest available focal spot, its size was
still a limiting factor in image quality for angiographic examination of the
arterial vessels in the equine foot.

The LM view images were very dark wiﬁrin ﬁre region of ﬁre dish]
phalanx, so the palmar digihl artery at ﬁre entrance to the solar canal was only
identiﬁed in four of ﬁre six feet. This dark shading was due to underexposure of

the ﬁrickest region, as seen in the inverted gray scale images. This problem was

 

' General Electric Stenoscop LE, Milwaukee, WI

80

partially overcome by surrounding the limb with lead gloves on the table to
absorb some of ﬁre background radiation. This caused a decrease in exposure of
ﬁre phototimer cells in ﬁre hble, resulting in an increased duration of exposure
time and improved detail in the distal phalanx on the ﬁnal images.

On the lateromedial view images, a vascular bed was observed at ﬁre
coronary region which was supplied by dorsal branches of the terminal arch and
dish] branches of ﬁre dorsal phalangeal artery. This ﬁnding suggests an
alternative route of vascular supply to the dorsal aspect of the hoof via ﬁre dorsal
phalangeal artery in case of a decrease in perfusion from the terminal arch.

The diameter of the palmar digihl artery and its major branches to ﬁre
foot were determined from measurements of the contrast enhanced lumen on
angiographic images, and corrected for magniﬁcation using ﬁre tohl
magniﬁcation factors determined in method 1. The same technique of image
measurement, using digihl calipers, was used to measure ﬁre marker images
and the angiographic images, so ﬁre source of error due to penumbra was
consistent. Had the calculated tohl magniﬁcation factors (meﬁrod 2) been used
instead, ﬁren ﬁre vessel diameter results would have been larger. In ﬁre human
medical literature on cardiac caﬁreterization and contrast angiography, reference
is made to a computer program ﬁrat determines a precise location of the edge of
a vessel for quanﬁhtive angiographyﬁ‘t‘;5 Use of ﬁris program would enhance

repeahbility of measurements of the same vessel, but the accuracy of these

81

measurements would still be uncertain since the actual size of ﬁre vessels is
unknown.

The digital vessel diameters were consistenﬁy larger in the lateral
recumbent position (LM images) than in the shnding position (PD images).
Repositioning of ﬁre limb from standing to lateral recumbency would have
caused a decrease in hydroshtic pressure which was evident as a decrease in
arterial perfusion pressure. The pump ﬂow rate was increased from 150 ml/ min
to approximately 175 ml/ min to return arterial perfusion pressure to 100 mm
Hg. In so doing, ﬁris increase in ﬂow rate may have caused vasodilation. Other
causes for the larger vessel diameter results in ﬁre LM images may have been
vascular reactions to ﬁre previous injection of contrast material from the PD
study or progressive failure of the in-vitro limb with increased time since
euthanasia of the horse. Further investigation is necessary to determine the
impact of each of ﬁrese factors on vessel size.

The results of ﬁris study determined that the diameters of the arterial
vessels were signiﬁcanﬁy smaller than those reported by Ackerman, et al.54 The
disparity between ﬁre results of ﬁris current model and ﬁre previously published
report may be due to several factors. This research project used in-vitro,
perfused limbs that were removed from the horses immediately following
euthanasia. The arterial perfusion pressure was mainhined at 100 mm Hg by a
pump and the perfusate was a heated, oxygenated Krebs-Henseleit solution

(Appendix A). While these parameters were designed to mimic physiologic

82

conditions, there were vast differences between ﬁris in-vitro model and the live
foot. Ackerman, et al, performed ﬁreir angiographic study on live horses and
ponies under general anesﬁresia, so systemic cardiovascular and neurological
responses contributed to ﬁreir results. The primary focus of ﬁris project was to
quantify ﬁre tohl magniﬁcation factors of the digital radiographic imaging
system then to use ﬁrese factors to correct image measurements for magniﬁcation
to determine actual vessel diameters. Ackerman, et al, did not discuss ﬁre effect
of image magniﬁcation, so ﬁre reported vessel sizes may have been direct
measurements from the radiographs resulting in larger diameter values than the
results reported in ﬁris ﬁresis. Without knowledge of the radiographic technique
used in the previous study, the magnitude of magniﬁcation cannot be
determined.

The advanhges of the model described in ﬁris thesis include ﬁre ability to
control the contents and temperature of ﬁre perfusate, the concentration of each
component, and ﬁre perfusion pressure. This model may be useful for research
in ﬁre role of vascular reactivity in equine laminitis. Several modiﬁcations to ﬁris
model should be considered to approximate more closely ﬁre conditions of an in-
vivo foot including, catheterizaﬁon of the palmar digital veins to monitor or
control venous pressure and weighing the limb before and after each
angiographic study to determine tissue perfusion rates and to investigate the

effect of interstitial edema formation.

CONCLUSION

This study investigated three major contributing factors to image
magniﬁcation in digihl radiography, including geometric, electronic and
photographic magniﬁcation. Variables of ﬁre radiographic imaging design ﬁrat
inﬂuenced each of ﬁrese components were focal-ﬁlm distance, image intensiﬁer
mode of function and laser printer format, respectively.

Two methods for: quantiﬁcation of the tohl magniﬁcation factors were
performed to determine ﬁre relationship between each of ﬁrese variables and
image magniﬁcation. The ﬁrst method compared ﬁre size of a mehllic marker
to the size of its radiographic image. It was demonstrated by ﬁre resth of ﬁris
procedure that the tohl magniﬁcation factors increased wiﬁr increasing focal-
ﬁlm dishnce, decreasing image intensiﬁer mode and decreasing printer format
(fewer images per sheet of ﬁlm). This technique was simple to perform and
would be convenient to repeat in a variety of clinical applications. The
limitation of ﬁris meﬁrod is the inaccuracy of deﬁning ﬁre edge of an image due
to the penumbra effect.

The second method calculated total magniﬁcation factors from ﬁre

product of geometric, electronic and photographic magniﬁcation. This

83

84
procedure was designed to provide comparable values to results of method 1

and to quantify the contributions of ﬁre individual components to tohl
magniﬁcation. The results of ﬁris second method produced smaller values for
tohl image magniﬁcation than the ﬁrst method. This underestimation of the
magniﬁcation factor was likely due to the intentional exclusion of ﬁre
contribution of penumbra to image magniﬁcation.

If it is the intention to measure an image and correct for magniﬁcation,
in order to determine an object's size, then a radiographic technique should be
selected to minimize image magniﬁcation. In so doing, ﬁre penumbra will be
minimized and deﬁnition of the image margin will be more distinct. This may
be accomplished in digihl radiography by using a short focal-ﬁlm dishnce,
assuming a ﬁxed focal-object dishnce. The largest image intensiﬁer mode
should be used and the greatest number of images possible should be printed
on each sheet of ﬁlm. However, if purpose of the radiographic study is for
subjective evaluation of small structures, then image magniﬁcation may
enhance visualization of ﬁre objects of interest, ﬁrerefore a compromise between
these techniques may be most useful. In eiﬁrer situation, the smallest available
focal spot should be used to produce the best image dehil by minimizing ﬁre
penumbra.

Digihl angiography of the equine dish] limb provided detailed images
wiﬁr adequate resolution to consistenﬁy identify the palmar digital artery and

its major branches to ﬁre foot. The terminal arch and its solar branches were

85
more easily seen on the palmarodorsal view while the lateromedial View was

more useful for examination of the bulbar and coronary arteries and the dorsal
branches of ﬁre terminal arch.

Diameters of the arterial vessels in these feet were determined, based on
measurements of their radiographic images and correction for image
magniﬁcation. The palmar digihl artery tapered from 2.28 mm, proximal to]
the bulbar artery, to 1.61 mm at the entrance to ﬁre solar canal of the distal
phalanx and 1.42 mm at ﬁre midsagithl point of the terminal arch. The values
reported for vessel diameters in ﬁris thesis were based on in-vitro perfused
equine limbs. This information may be applied to further research in normal

and pathological vascular reactivity.

APPENDIX A

APPENDD( A

Directions for preparation of Krebs Henseleit solution

To prepare 4.0 L Krebs Henseleit solution:

 

m 4 stock wlrmm:
1) KCL 28.034 g/ L
2) CaClz-ZHZO 30.58 g/ L
3) KHzPO4 12.63 g/ L
4) MgSO4-7H20 23.464 g/ L

2221:

Mix 1) 50 ml KC] solution 4.70 mM
2) 50 ml KHzPO4solutiorr 1.16 mM

3) 50 ml Mg504-7HzO solution 1.19 mM
In 4) 3000 ml P120

@3122:

Add 1) NaCl 27.68 g
2) Dextrose 8.43 g
3)“ NaHCO3 8.40 g

St_ep 3:

Add 1) 50nﬁCaClr2H20 2.6 mM
2) 780 ml H20

86

APPENDIX B

APPENDIX B

Table 7 A metallic marker was radiographed at several focal-film distances (FFD), using three
different image intensiﬁer tube modes (II mode) and printed at four formats. Each image was
measured ﬁrree times wiﬁr calipers.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FFDcm II mode = format rep] mar rep2 mar rep3 mar avg mm St Dev X
74.5 9 12 8.44 8.45 8.41 8.83 0.36083
79.5 9 12 8.74 8.78 8.74 9.18 0.31969
84.5 9 12 9.31 9.05 9.56 9.54 0.21119
87.5 9 12 9.49 9.46 9.66 9.9 0.37175
89.5 9 12 9.82 9.71 9.75 10.79 0.99629
94.5 9 12 10.39 10.26 10.21 11.79 1.02596
74.5 6 12 12.16 12.04 12.19 12.7467 0.45206
79.5 6 12 12.82 12.77 13.39 13.2767 0.37985
84.5 6 12 13.26 13.28 13.83 13.7133 0.35612
87.5 6 12 13.75 13.8 13.98 14.21 0.41215
89.5 6 12 14.13 14.22 14.14 15.28 1.21461
94.5 6 12 14.75 14.95 14.98 16.4767 1.25974
74.5 4.5 12 16.96 16.82 16.7 17.7267 0.62872
79.5 4.5 12 17.72 17.58 18.67 18.47 0.6005
84.5 4.5 12 18.5 18.4 19.17 19.1067 0.46507
87.5 4.5 12 19.19 19.35 19.62 19.72 0.47378
89.5 4.5 12 19.63 19.76 20.61 17.0067 4.22196
94.5 4.5 12 20.34 20.58 20.79 14.3567 4.23867
74.5 9 6 11.05 11.18 11.12 11.7 0.53907
79.5 9 6 1 1 .68 1 1 .71 1 1.73 1 2.2733 0.45021
84.5 9 6 12.37 12.41 12.44 12.7667 032252
87.5 9 6 12.77 12.91 12.86 13.21 0.38131
89.5 9 6 13.16 13.23 13.04 14.2533 1.18506
94.5 9 6 13.7 13.75 13.53 15.4067 1.24208
74.5 6 6 15.9 16.29 15.7 16.8533 0.88909
79.5 6 6 16.62 16.62 16.78 17.66 0.74422
84.5 6 6 18.04 18.21 17.74 18.47 0.42575
87.5 6 6 1832 18.64 18.62 19.09 0.64565
89.5 6 6 19.05 19.04 19.09 20.3767 1.32152
94.5 6 6 19.9 19.77 19.85 21.7567 137609
74.5 4.5 6 22.18 22.21 22.24 23.53 1.26415
79.5 4.5 6 23.19 23.31 23.17 24.82 1.20136
84.5 4.5 6 25.22 25.07 25.03 25.8567 0.46162
87.5 4.5 6 26.05 26.1 25.75 26.6567 0.68878
89.5 4.5 6 26.3 26.59 26.3 22.2533 6.67801
94.5 4.5 6 27.62 27.38 27.49 17.9567 6.83697

 

 

 

 

 

 

 

 

87

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 7 continued
FFD an mode format rep] mm rep2 mm rep3 mm avg mm St Dev X
74.5 9 4 12.84 12.85 12.83 13.4467 0.51097
79.5 9 4 13.41 13.46 13.71 14.09 0.55522
84.5 9 4 14.09 14.28 14.29 14.56 0.33297
87.5 9 4 14.77 14.6 14.67 15.0533 0.36591
89.5 9 4 14.82 15.04 15.2 16.3133 1.61093
94.5 9 4 15.57 15.94 15.68 17.8367 1.63905
74.5 6 4 18.55 18.68 18.49 19.5267 0.85869
79.5 6 4 1939 19.67 19.36 20.4667 0.81757
84.5 6 4 20.64 20.77 20.53 21.3033 0.51655
87.5 6 4 21.37 21.28 21.17 22.0733 0.65657
89.5 6 4 21.9 21.74 22.02 23.5 1.57956
94.5 6 4 22.95 22.79 22.9 25.3533 1.85311
74.5 4.5 4 25.65 25.73 25.75 27.2633 1.24478
79.5 4.5 4 27.46 27.17 27.18 28.6067 0.90779
84.5 4.5 4 28.68 28.85 28.93 29.4967 0.60599
87.5 4.5 4 29.68 29.66 29.8 30.56 0.94435
89.5 4.5 4 30.13 30.55 30.58 29.3 2.50691
94.5 4.5 4 31.87 31.57 31.68 28.35 2.55198
74.5 9 1 25.9 25.8 25.86 27.2867 1.13494
7'95 9 1 27.28 27.19 27.26 28.5567 0.99587
84.5 9 1 28.68 28.83 28.7 29.6467 0.76474
87.5 9 1 29.71 29.67 29.64 30.74 0.92833
89.5 9 1 30.55 30.64 30.48 33.22 2.83792
94.5 9 1 31.96 31.9 31.88 36.0633 3.00657
74.5 6 1 37.15 37.24 37.28 39.3067 1.86036
79.5 6 1 39.08 39.66 39.45 41.3767 1.76129
84.5 6 1 41 .69 41 .68 41 .8 43.17 1.1388
87.5 6 1 43.36 43.52 43.14 44.72 1.23039
89.5 6 1 44.46 44.11 44.53 47.6333 3.25(B3
94.5 6 1 46.34 46.74 46.68 51.1233 3.5742]
74.5 4.5 1 52.1 52.29 52.3 55.18 2.62284
79.5 4.5 1 54.93 54.85 55.06 581333 2.36342
84.5 4.5 1 58.51 58.78 58.62 60.5067 1.57142
87.5 4.5 1 60.66 60.45 60.54 62.6833 1.8036
89.5 4.5 1 62.35 62.32 61.89 63.695 26.0659
94.5 4.5 1 65.04 64.79 65.22 65.04 26.5993

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

89
Shtisﬁcal analysis of measured Tohl Magniﬁcation Factors (T MFm)

General Linear models procedure, multiple regression analysis

Output of Shtistix 3.1 Analytical Software

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT'S T P
CONSTANT 3.9290 4.3365E-01 9.06 0.0000
IIMODE -2.9357E-01 6.4113E-02 -4.58 0.0000
CASES INCLUDED 72 MISSING CASES 0

DEGREES OF FREEDOM 70

OVERALL F 20.97 P VAIUE 0.0000

ADJUSTED R SQUARED 0.2195

R SQUARED 0.2305

RESID. MEAN SQUARE 1.036

UNVEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT’S T P
CONSTANT 3.1406 1.7378E-01 ' 18.07 0.0000
FORMAT -1.9474E-01 2.4763E-02 -7.86 0.0000
CASES INCLUDED 72 MISSING CASES 0

DEGREES OF FREEDOM 70

OVERALL F 61.84 P VALUE 0.0000

ADJUSTED R SQUARED 0.4615

R SQUARED 0.4691

RESID. MEAN SQUARE 7.147E-01

UNUEIGHTED LEAST SQUARES LINEAR REGRESSION OF THEM

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT’S T P
CONSTANT 1.7944E-01 1.7656 0.10 0.9193
FFD 2.1663E-02 2.0711E-02 1.05 0.2992
.CASES INCLUDED 72 MISSING CASES 0

DEGREES OF FREEDOM 70 "

OVERALL P 1.094 P VALUE 0.2992

ADJUSTED R SQUARED 0.0013

R SQUARED 0.0154

RESID. MEAN SQUARE 1.325

PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 2.0877

FFD 2.1663E-02
IIMODE -2.9357E-01
CASES INCLUDED 72
DEGREES OF FREEDOM 69
OVERALL F 11.25
ADJUSTED R SQUARED 0.2240
R SQUARED 0.2459

RESID. MEAN SQUARE 1.030

STD ERROR STUDENT'S T
1.6109 1.30
1.8256E-02 1.19
6.3927E-02 -4.59

MISSING CASES 0

P VALUE 0.0001

UNVEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT’S T
CONSTANT 1.2992 1.2946 1.00
FFD 2.1663E-02 1.5094E-02 1.44
FORMAT -1.9474E-01 2.4578E-02 -7.92
CASES INCLUDED 72 MISSING CASES 0
DEGREES OF FREEDOM 69

OVERALL F 32.42 P VALUE 0.0000
ADJUSTED R SQUARED 0 . 4 695

R SQUARED 0.4845

RESID. MEAN SQUARE 7.040E-01

UNWEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 5.0488

FORMAT -1.9474E-01
IIMODE -2.9357E-01
CASES INCLUDED 72
DEGREES OF FREEDOM 69
OVERALL F 80.33 '

ADJUSTED R SQUARED 0.6908
R SQUARED 0.6996
RESID. MEAN SQUARE

STD ERROR STUDENT'S T
2.9347E-01 17.20
1.8763E-02 -10.38
4.0350E-02 -7.28

MISSING CASES 0

P VALUE 0.0000

4.103E-01

0.0000

0.0000

0.0000
0.0000
0.0000

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT’S T
CONSTANT 3.2074 ' 1.0032 3.20
FFD 2.1663E-02 1.13068-02 1.92
FORMAT -1.9474E-01 1.84108-02 -10.58
IIMODE ~2.9357E-01 3.9591E-02 -7.42
CASES INCLUDED 72 MISSING CASES 0
DEGREES OF FREEDOM 68

OVERALL F 56.85 P VALUE 0.0000
ADJUSTED R SQUARED 0.7024

R SQUARED 0.7149

RESID. MEAN SQUARE 3.950E-01

91

UNWEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT'S T
CONSTANT 1.3387 3.5777 0.37
FFD 5.6170E-02 4.1766E-02 1.34
FORMAT -1.8977E-01 2.3561E-01 -0.81
IIMODE -1.7423E-01 4.9250E-01 -0.35
I1 -3.3306E-03 5.7425E-03 -0.58
12 -2.23628-03 2.6702E-03 -0.84
I3 2.8479E-02 9.3502E-03 3.05
CASES INCLUDED 72 MISSING CASES 0
DEGREES OF FREEDOM 65

OVERALL F ' 33.20 P VALUE 0.0000
ADJUSTED R SQUARED 0.7313

R SQUARED 0.7540

RESID. MEAN SQUARE 3.566E-01

UNWEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 3.1739

FFD 3.45223-02
FORMAT -1.8977E-01
IIMODE -4.s7333-01
12 -2.2362E-03
I3 2.84798-02
cases INCLUDED '72
DEGREES OF FREEDOM 55
OVERALL F 40.13

ADJUSTED R SQUARED 0 . 734 0
R SQUARED 0.7527
RESID. MEAN SQUARE

STD ERROR
1.6452

1.8645E-02
2.3443E-01
6.5288E-02
2.6567E-03
9.3031E-03

STUDENT’S T

MISSING CASES 0

P VALUE

3.531E-01

0.0000

0.0000
0.0000

0.0033

0.0032

PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 4.2718

FFD 2.1663E-O2
FORMAT -3.7985E-01
IIMODE -4.5733E-01
I3 2.8479E-02
CASES INCLUDED 72
DEGREES OF FREEDOM 67
OVERALL F 50.26
ADJUSTED R SQUARED 0.7351
R SQUARED 0.7501
RESID. MEAN SQUARE

92.

STD ERROR

1.0666E-02
6.2788E-02
6.5146E-02
9.2829E-03

STUDENT’S T

MISSING CASES 0

P VALUE

3.515E-01

0.0000

UNWEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

STD ERROR

12.156

2.8936E-01
6.32628-02
6.5637E-02
9.35298-03
1.71508-03

STUDENT'S T

0.07
~6.00
-6.97

3.04

0.01

MISSING CASES 0

P VALUE

3.568E-01

0.0000

UNHEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 4.3630

FFD 1.9487E-02
FORMAT -3.798SE-01
IIMODE -4.s733E-01
I3 2.8479E-02
FFDZ 1.2908E-OS
CASES INCLUDED 72
DEGREES OF FREEDOM 66
OVERALL F 39.61
ANUSTED R SQUARED 0.7311
R SQUARED 0.7501
RESID. MEAN SQUARE
PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 7.2986

FFD 2.1663E-02
FORMAT -3.7985E-01
IIMODE -1.4139

I3 2.6479E-02
IIMODE2 6.97532-02
CASES INCLUDED 72
DEGREES OF FREEDOM 66
OVERALL F 43.25

ADJUSTED R SQUARED 0.7485
R SQUARED 0.7662
RESID. MEAN SQUARE

STD ERROR
1.7253

1.03948-02
6.1186E-02
4.5275E-01
9.0460E-03
3.2687E-02

STUDENT’S T

4.23
2.08
-6.21
-3.12
3.15
2.13

MISSING CASES 0

P VALUE

3.338E-01

0.0000

0.0031

0.0000
0.0033
0.9940

93

STUDENT'S T

MISSING CASES 0

PREDICTOR

VARIABLES COEFFICIENT STD ERROR
CONSTANT 5.3298 4.9873E-01
FFD 2.1663E-02 5.2213E-03
FORMAT -8.8108E-01 4.6055E-02
IIMODE -4.5733E-01 3.1891E-02
I3 2.8479E-02 4.5443E-03
FORMAT2 3.7038E-02 2.5344E-03
CASES INCLUDED 72

DEGREES OF FREEDOM 66

OVERALL F 210.5 P VALUE
ADJUSTED R SQUARED 0.9365

R SQUARED 0.9410

RESID. MEAN SQUARE 8.424E-02

0.0000

UNWEIGNTED LEAST SQUARES LINEAR REGRESSION OF TMFM

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT’S T
CONSTANT 8.3565 7.4703E-01 11.19
FFD 2.1663E-02 4.4847E-03 4.83
FORMAT -8.8108E-01 3.9558E-02 -22.27
IIMODE -1.4139 1.9535E-01 -7.24
I3 2.8479E-02 3.9032E-03 7.30
IIMODE2 6.9753E-02 1.4104E-02 4.95
FORMATZ 3.7038E-02 2.1768E-03 17.01
CASES INCLUDED r72 MISSING CASES 0
DEGREES OF FREEDOM

OVERALL F 241.9 P VALUE 0.0000
ADJUSTED R SQUARED 0.9532

R SQUARED 0.9571

RESID. MEAN SQUARE 6.215E-02

UNWEIGHTED LEAST SQUARES LINEAR REGRESSION OF TMFM

STUDENT’S T

-9.35
9.43
6.39
8.27

-6.68

MISSING CASES 0

PREDICTOR

VARIABLES COEFFICIENT STD ERROR
CONSTANT 8.9680 5.8521E-01
FFD 2.1663E-02 3.4699E-03
FORMAT -1.5701 1.0764E-01
IIMODE -1.4139 1.5115E-01
I3 2.8479E-02 3.0200E-03
IIMODE2 6.9753E-02 1.0912E-02
FORMAT2 1.8996E-01 2.2966E-02
FORMAT3 -8.3247E-03 1.2468E-03
CASES INCLUDED 72

DEGREES OF FREEDOM 64

OVERALL F 352.7 P VALUE
ADJUSTED R SQUARED 0.9720

R SQUARED 0.9747

RESID. MEAN SQUARE 3.720E-02

0.0000

0.0001
0.0000
0.0000
0.0000
0.0000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

0.0000
0.0000

94

RANKITS VS RESID

 

RANKITS
3.0 +
+ +
+ +
+2+
1.0 + ++32
+54
263
++45
82
-1.0 + 6+
4
2
2
-3.0 +
-+ --------- + --------- + --------- + --------- +-
-4.0 -l.0 2.0 5.0 8.0

RESID X 10E-1
APPROX. WILK-SHAPIRO 0.9472 72 CASES PLOTTED

95

Table 8 A linear function, defined by linear regression analysis, was used to calculate ﬁtted
values for TMFm at each focal-film distance (FFD), image intensifier tube mode (II mode) and
printer format.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FFD IIMODE FORMAT TMFM FITTED RESID
7450 9.00 12.00 0.70 0.71 -0.01
79.50 9.00 12.00 0.73 0.82 .009
84.50 9.00 1200 0.77 0.93 -0.16
8750 9.00 12.00 0.79 0.99 .020
8950 9.00 12.00 0.81 1.04 .023
9450 9.00 12.00 056 1.14 -0.28
74.50 6.00 12.00 1.01 0.79 0.22
79.50 6.00 12.00 1.08 0.90 0.18
84.50 6.00 , 12.00 1.12 1.00 0.12
87.50 6.00 1200 1.15 1.07 0.08
89.50 6.00 12.00 1.18 1.11 0.07
94.50 6.00 12.00 1.24 1.22 0.02
74.50 4.50 12.00 1.40 1.30 0.10
79.50 4.50 1200 1.50 1.41 0.09
84.50 450 12.00 1.55 1.51 0.04
87.50 450 12.00 1.61 1.58 0.03
89.50 4.50 12.00 1.66 1.62 0.04
9450 450 12.00 1.71 1.73 .002
7450 9.00 6.00 0.93 0.66 0.27
7950 9.00 6.00 0.97 0.77 0.20
8450 9.00 6.00 1.08 0.88 0.15
8750 9.00 6.00 1.05 0.95 0.10
8950 9.00 6.00 1.09 0.99 0.10
9450 9.00 6.00 1.14 1.10 0.04
7450 6.00 6.00 133 1.25 0.08
79.50 6.00 6.00 139 1.36 0.03
8450 6.00 6.00 1.50 1.47 0.03
8750 6.00 6.00 1.54 1.54 0.00
8950 6.00 6.00 1.59 1.58 0.01
94.50 6.00 6.00 1.65 1.69 .004
74.50 4.50 6.00 1.85 2.02 .0.17
7950 4.50 6.00 1.93 2.13 .020
8450 450 6.00 2.09 2.24 .0.15
87.50 450 6.00 2.16 2.30 -0.14
8950 450 6.00 2.20 2.35 -0.15
94.50 450 6.00 229 2.45 -O.16
7450 9.00 4.00 1.07 0.76 0.31
7950 9.00 4.00 1.13 0.87 0.26
8450 9.00 4.00 1.18 0.97 0.21
8750 9.00 4.00 1.22 1.04 0.18
89.50 9.00 4.00 1.25 1.08 0.17
94.50 9.00 4.00 1.31 1.19 0.12
74.50 6.00 4.00 1.54 1.52 0.02
79.50 6.00 4.00 1.62 1.63 0.01
84.50 6.00 4.00 1.72 1.74 .002
87.50 6.00 4.00 1.77 1.80 -003
89.50 6.00 4.00 1.82 1.84 .002
94.50 6.00 4.00 1.90 1.95 .005
74.50 4.50 4.00 2.14 2.37 .023
79.50 4.50 4.00 2.27 2.48 .021
84.50 4.50 4.00 2.40 2.59 0.19
87.50 4.50 4.00 2.47 2.65 -0.18

 

 

 

 

 

 

 

 

 

 

 

 

 

96
Table 8 continued

TMFM FITTED RESID
2.53 2.70 --0.17
2.64 2.80 -0.16
2.15 2.37 -0.22
2.27 2.48 -0.21
2.39 2.59 -0.20
2.47 2.66 0.19
2.54 2.70 -0.16
2.65 2.81 -0.16

3.39 -0.29
3.28 3.50 -0.22

3.61 -0.14

3.67 -0.06

3.72 -0.03

3.83 0.05

4.37 —0.02

4.48 0.09

4 .59

4.65

4.70

4.80

 

97

 

 

 

 

 

2.5 _A
g .5
'5 1 '
a: 0.5 -
o I I I I I I I I I I I I I I I I I 1 I I
“2 “'2 “I “2 "2 “2 “'2 ‘0. “'2 “2 “2
E E2 82 8 36 S 8 8 8 a a
focal-film distance (cm)

Figure 32 Fitted values for total magnification factors (TMFm), averaged over all II modes and
formats, increased gradually with increasing FFD.

 

 

 

 

 

 

 

 

3
__. +———Of+
32.5
'51 7
ans
“I I IIII I I I I I I ITI I III I1
“I “2 “2 “2 "2 "2 “'5 “2
E :2 :2 3 8 8‘ 8'

00
focal-film distance (cm)

Figure 33 Fitted values for total magnification factors (TMFm), averaged over all formats were
higher for smaller 11 modes, but as FFD increased, the trend of increasing magniﬁcation factors was
the same for each [I mode.

 

 

 

 

 

 

 

 

 

 

3.5 + W I
a 3 +1501
Eu +4.01
2 +6401
1: _. m— :: +1201

$1.: + __*______*___x_ ——.x

0.5

o I I I I I I I I I I f f ﬂ I I I7 I I I I I

"2 "a "2 “'2 m "2 19 m. "2 we "a

E R K 8 g 3 8 8 8 S 34

focal-fﬂm distance (cm)

Figure 34 Fitted values for total magnification factors (TMFm), averaged over all II modes, were
higher for formats with fewer images per Sheet of film, but as FFD increased, the trend of increasing
magnification factors was the same for each format.

98

   

2..

l

l

 

o
clan-A

 

6 inch 9 inch
image intensifier mode

average fitted TMFm
G .4

8
‘3"
D‘

Figure 35 Fitted values for total magniﬁcation factors (TMFm), averaged over all FFD's and formats,
decreased steadily with increasing II mode.

 

 

 

 

 

a 5

E 4
a 3 +1.01
é +401
a 2 +901
an 1 ~34 +1201
E o I I‘ I

4.5 inch 6 inch 9 inch

image intensifier mode

Figure 36 Fitted values for total magniﬁcation factors (TMFm), averaged over all FFD's, were higher

forformatswithfewerimagespersheetoffilm. AsIImodeincreased,fitted'I'MFmdecreasedat
everyformat.

 

+ 74.5 cm
—I— 79.5 cm
+ 84.5 cm
—M-— 87.5 cm
0 -O— 89.5 cm

-O—94.S cm
4.5 inch 6 inch 9 inch

 

 

average fitted TMFm

 

 

 

image intensifier mode

Figure 37 Fitted values for total magnification factors (Mm), averaged over all formats, were
higher for longer FFD's, but as 11 mode increased, the trend of decreasing magnification was the
same for each FFD.

99

15

3-

average fitted TMFm
N

 

O

 

I I I

4:01 . 6:01 12:01

printer format

§

Figure 38 Fitted values for total magnification factors (TMFm), averaged over all FFD' s and II
modes, decreased as format increased.

 

 

 

 

 

 

4

8 +746 cm
E 3 .1 +795 em
.3 +845 cm
g z . +875 cm
6:: +895 cm
56 1 . —o—s4.6 an
a o I I I

Figure 39 Fitted values for total magnification factors (TMFm), averaged over all 11 modes, were

higher for longer FFD's, but as format increased, the trend of decreasing magnification factors was
the same for each FFD.

I +4.5 inch
+6 inch
\ +9 inch

I I

12:01

  
   

 
 

 

 

 

 

average fitted TMFm
o a N w s- m

o‘lpr'inter “th

Figure 40 Fitted values for total magnification factors (TMFm), averaged over all FFD‘s, decreased
as the number of images per sheet of film (format) increased.

 

APPENDIX C

APPENDIX C

Table 9 Magnification factors were calculated for individual components of the digital
radiographic imaging system, at each focal-film distance, image intensifier mode and printer
format.

FOCAL-FILM DISTANCE MAGNIFICATION FACTOR
25 cm 1.50
30 cm 1.60
35 cm 1.70
38 cm 1.77
40 cm 1.80
45 cm 1.90

IMAGE INTENSIFIER MODE
9 INCH Me=0.10
6 INCH ~ Me=0.15
4.5 INCH Me=0.20

PRINTER FORMAT
1:1
4:1
6:1
12:1

 

100

101
Statistical analysis of Delta (TMFm-TMFc)

Descriptive Statistics

Output of Statistix 3.1 Analytical Software

Table 10 There was some variation in the difference between measured and calculated total
magnification factors (delta) between the radiographic technique variables.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

VARIABLE MEAN DELTA SAMPLE SIZE VARIANCE STAND DEV

FFD

74.5 0.1767 12 0.01444 0.120

79.5 0.1667 12 0.01272 0.113

84.5 0.1717 12 0.01654 .0129

87.5 0.1608 12 0.01604 0.127

89.5 0.1825 12 0.01873 0.137

94.5 0.1700 12 0.01818 0.135

TOTAL 0.1714 72 X=0.1 27
II MODE

4.5 0.2771 24 0.02151 0.147

6 0.1 167 24 0.003736 0.061

9 0.1204 24 0.003639 0.060

TOTAL 0.1714 72 X=0.089
FORMAT

1 0.3200 18 0.02193 0.148

4 0.1333 18 0.00380 0.062

6 0.1267 18 0.003153 0.056

12 0.1056 18 0.002238 0.047

TOTAL 0.1714 72 X=0.078

 

 

 

 

 

 

102
Statistical analysis of Delta (TMFm-TMFc)

General Linear models proceure, multiple regression analysis

Output of Statistix 3.1 Analytical Software

PREDICTOR '

VARIABLES COEFFICIENT STD ERROR STUDENT'S T P
CONSTANT 1.7772E-01 - 1.8932E-01 0.94 0.3511
FFD -7.4434E-05 2.2207E-03 -0.03 0.9734
CASES INCLUDED 72 MISSING CASES 0

DEGREES OF FREEDOM 70

OVERALL F 1.124E-03 P VALUE 0.9734

ADJUSTED R SQUARED -0.0143

R SQUARED 0.0000

RESID. MEAN SQUARE 1.524E-02

UNWEIGMTED LEAST SQUARES LINEAR REGRESSION 0F DELTA

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT’S T P
CONSTANT 3.6420E-01 4.6812E-02 7.78 0.0000
IIMODE -2.9663E-02 6.9209E-03 '4.29 0.0001
CASES INCLUDED -72 MISSING CASES 0

DEGREES OF FREEDOM 70

OVERALL F 18.37 P VALUE 0.0001

ADJUSTED R SQUARED 0.1966

R SQUARED 0.2079

RESID. MEAN SQUARE 1.207E-02

UNWEIGHTED LEAST SQUARES LINEAR REGRESSION OF DELTA

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT'S T p
CONSTANT 2.6569E-01 2.13238-02 12.46 0.0000
FORMAT -1.6401E-02 3.03843-03 -s.40 0.0000
CASES INCLUDED 72 MISSING CASES 0

DEGREES OF FREEDOM 70

OVERALL F 29.14 P VALUE 0.0000

ADJUSTED R SQUARED 0.2838 ‘

R SQUARED 0.2939-

RESID. MEAN SQUARE 1.076E-02

103

PREDICTOR

VARIABLES COEFFICIENT STD ERROR STUDENT'S-T

----------- .b-------- -------:- 2.11

CONSTANT 3.7052E-01 1.7565E 01

FFD -7.4434E-05 1.99078-03 -0.04

IIMODE -2.9663E-02 6.97088-03 -4.26

CASES INCLUDED 72 MISSING CASES 0
REES OF FREEDOM 69

UVERALL F 9.054 P VALUE 0.0004

ADJUSTED R SQUARED 0.1849

R SQUARED 0.2079

RESID. MEAN SQUARE 1.225E-02

UNNEIGNTED LEAST SQUARES LINEAR REGRESSION OF DELTA

PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 2.7202E-01
FFD -7.4434E-05
FORMAT -1.6401E-02
CASES INCLUDED 72
DEGREES OF FREEDOM 69
OVERALL F 14.36

ADJUSTED R SQUARED 0.2735
R SQUARED 0.2939
RESID. MEAN SQUARE

STD ERROR STUDENT’S T
1.6119E-01 1.69
1.8795E-03 -0.04
3.0603E-03 -5.36

MISSING CASES 0

P VALUE 0.0000

1.092E-02

UNVEIGETED LEAST SQUARES LINEAR REGRESSION OF DELTA

PREDICTOR

VARIABLES COEFFICIENT
CONSTANT 4.5850E-01
IIMODE -2.9663E-02
FORMAT -1.6401E-02
CASES INCLUDED 72
DEGREES OF FREEDOM 69
OVERALL F 34.75
ADJUSTED R SQUARED 0.4873
R SQUARED 0.5018

RESID. MEAN SQUARE

STD ERROR STUDENT’S T
4.0209E-02 11.40
5.5284E-03 -5.37
2.5707E-03 -6.38

MISSING CASES 0

P VALUE 0.0000

7.702E-03

0.9703
0.0001

0.0000

0.0000

FREDICTOR
VARIABLES COEFFICIENT STD ERROR
CONSTANT 4.6483E-01 1.4111E-01
FFD -7.4434E-05 1.59038-03
IIMODE -2.9663E-02 5.5688E-03
FORMAT -1.64OIE-02 2.5894E-03
CASES INCLUDED 72

DEGREES OF FREEDOM 68

OVERALL F 22.83 P VALUE
ADJUSTED R SQUARED 0.4798

R SQUARED 0.5018

RESID. MEAN SQUARE 7.8158-03

 

104

. STUDENT’S T

MISSING CASES 0

0.0000

UNWEIGHTED LEAST SQUARES LINEAR REGRESSION OF DELTA

PREDICTOR

VARIABLES COEFFICIENT STD ERROR
CONSTANT 3.8654E-01 5.2278E-01
FFD 2.1311E-03 6.1028E-03
IIMODE -2.9385E-02 7.1964E-02
FORMAT -2.2087E-02 3.4428E-02
I1 -2.0088E-04 8.3909E-04
12 -1.5649E-04 3.9017E-04
I3 2.9212E-03 1.3663E-03
CASES INCLUDED 72

DEGREES OF FREEDOM 65

OVERALL F 12.51 P VALUE
ADJUSTED R SQUARED 0.4932

R SQUARED 0.5360

RESID. MEAN SQUARE 7.615E-03

STUDENT’S T

MISSING CASES 0

0.0000

UNVEIGHTED LEAST SQUARES LINEAR REGRESSION OF DELTA

PREDICTOR

VARIABLES COEFFICIENT STD ERROR
CONSTANT 5.7044E-01 9.2441E-02
IIMODE -4.6460E-02 9.4598E-03
FORMAT -3.5388E-02 9.1175E-03
I3 2.9212E-03 1.3480E-03
FFD2 -3.8002E-07 9.1797E-06
CASES INCLUDED 72 MISSING
DEGREES OF FREEDOM 67

OVERALL F 19.23 P VALUE
ADJUSTED R SQUARED 0.5066

R SQUARED 0.5344

RESID. MEAN SQUARE 7.412E-03

 

STUDENT'S T

CASES 0

0.0000

0.0000
0.0000

0.0363

0.9671

105

STUDENT’S T

6.63

MISSING CASES 0

‘PREDICTOR

VARIABLES COEFFICIENT STD ERROR
CONSTANT 6.7826E-01 5.2462E-02
IIMODE -4.6460E-02 7.3531E-03
FORMAT -8.77SOE-02 1.0619E-02
I3 2.9212E-03 1.0478E-03
FORMAT2 3.8715E-03 5.8435E-04
CASES INCLUDED 72

DEGREES OF FREEDOM 67

OVERALL F 42.80 P VALUE
ADJUSTED R SQUARED 0.7019

R SQUARED 0.7187

RESID. MEAN SQUARE 4.478E-03

0.0000

UNNEIGRTED LEAST SQUARES LINEAR REGRESSION OF DELTA

PREDICTOR

VARIABLES COEFFICIENT STD ERROR
CONSTANT 1.7216 1.0963E-01
IIMODE -3.7619E-01 3.3336E-02
FORMAT -8.7780E-02 6.7504E-03
I3 2.9212E-O3 6.6607E-04
FORMAT2 3.8715E-03 3.7147E-04
IIMODE2 2.4043E-02 2.4068E-03
CASES INCLUDED 72

DEGREES OF FREEDOM 66

OVERALL F 104.7 P VALUE
ADJUSTED R SQUARED 0.8795

R SQUARED 0.8880

RESID. MEAN SQUARE 1.810E-03

STUDENT'S T

MISSING CASES 0

0.0000

UNNEIGETED LEAST SQUARES LINEAR REGRESSION OF DELTA

PREDICTOR
VARIABLES
CONSTANT
IIMODE
FORMAT

I3
FORMAT2
IIMODE2
FORMAT3

COEFFICIENT

1.8004
-3.7619E-01

~1.7657E-01.

2.9212E-03
2.3579E-02
2.4043E-02
'1.0728E-03

CASES INCLUDED
DEGREES OF FREEDOM

OVERALL F 114.9 P VALUE
ADJUSTED R SQUARED 0.9059

R SQUARED 0.9138

RESID. MEAN SQUARE 1.414E-03

STD ERROR
9.8532E-02
2.9465E-02
2.0983E-02
5.8872E-04
4.4769E-03
2.1273E-03
2.4306E-04

STUDENT’S T
18.27
-12.77
-8.41
4.96
5.27
11.30
-4.41

MISSING CASES 0

0.0000

0.0000
0.0000
0.0000
0.0069
0.0000

0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

P
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

106

STATISTIX 3.1

RANKITS VS RESID2

RANKITS
3.0 +
'I- 'I-
+ +
+ +2
1.0 'I' 6 +
28
227
4322
424
“1.0 '1‘ 3+3
2 2
2
2
'3.0 +
-+ --------- 'I' --------- + --------- 'I' --------- +-
‘0.8 '0.3 0.2 0.7 1.2

RESIDZ X 10E-1
APPROX. WILK-SHAPIRO 0.9720 72 CASES PLOTTED

107

 

 

 

5 0.2

7 ‘0‘ 2 ¢ # 4. 4

3 E 0.15 L

0

é a 0 1 -

0

gEDW«

0

g .

CI 0 r I I I 1
74.5 79.5 84.5 87.5 89.5 94.5

focal-fﬂm distance (cm)

Figure 41 Fitted delta (difference between measured and calculated magnification factors),
averaged over all II modes and formats, was the same at every FFD.

 

+4.5 inch
+6 inch

+ 9 inch

   

average fitted delta
(TMFm-MC)

 

0 f I 1 ' '

74.5 79.5 fag-film diggce (Ct: '5 94.5

Figure 42 Fitted delta, averaged over all formats, was highest at the 4.5 inch II mode.

 

 

(w) 991121819 tum-rm;
9°76 9‘68 9'88 5‘78 5'66 9°74

1021+ SI
m=9+ (I:
IO-‘Ir—I— a
tin-O— &
O-

E

E

 

Figure 43 Fitted delta, averaged over all II modes, was highest at the 1:1 format.

108

 

 

 

g 2 0.3

.8 g 0.25 -

'3 0.2 1

g E 0. -

0.90 0.1 -1 ' I

a E 0.05 1

0

g 0 I I
4.5 inch 6 inch 9 inch

image intensifier mode

Figure 44 Fitted delta, averaged over all formats and FFD's, was highest at the 4.5 inch II mode.

 

 

 

 

 

 

3 0.6
E 770.4
'5 E +1:01
g 0.3

A +4:01
a": in _,
5° 5 + 6:01
v 0.1 +1201
D
U o I I '

4.5 inch 6 inch

9 inch
image intensifier mode

Figure 45 Fitted delta, averaged over all FFD's, was highest at the 1:1 format.

 

 

 

 

 

 

 

% A03 . +74.Scrn
z 0.2 : +79.Scrn
g +84.Scm
a £01 I +8756...
6 . "

no +89.Scrn
E E— 0 ' +94.Scn1
> I

G

4.5 inch 6 inch , 9 inch

image intensifier mode

Figure 46 Fitted delta, averaged over all formats, was the same at every FFD.

109

 

I I

12:01

 

average fitted delta
(TMFm-Md
a 8 8 a
at
I

’8'

'01 printer format 6‘01

Figure 47 Fitted delta, averaged over all 11 modes and FFD's, was greatest at the 1:1 format.

 

 

 

 

 

 

 

 

g A 00.3 q +745
19 0.25 - +795
'8 0.2 - +845
3.": 0 .15 4 +875
- E + +

. U o 1 - H 89.5
g 0.05 - 1+9“
3 o : I I ﬂ

1:01 4:01 printer format 6:01 12:01

Figure 50 Fitted delta, averaged over all II modes, was the same at every FFD.

0.5

0.4

0.3 + 4.5 inch

02 +6 inch

0.1 4‘ + 9 inch
ﬂ

j

 

 

('IWm-TMFC)

 

 

 

 

average fitted delta

1:01 4. 01 printer forma6: 01 12:01

Figure 49 Fitted delta, averaged over all FFD's, was highest at the 4.5 inch 11 mode.

APPENDIX D

 

APPENDIX D

Table 11 Arterial vessels of six in-vitro equine feet were measured on digital angiograms (image
size) in the palmarodorsal (PD) and lateromedial (LM) views. Measurements were corrected for

magnification to determine the diameter of each vessel (vessel size).

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

PALMAR DIGITAL A PROXIMAL TO BULBAR A

HORSE PD IMAGE SIZE (mm) PD VESSEL SIZE (mm) LM IMAGE SIZE (mm) LM VESSEL SIZE (mm)
1 2.27 2.16 4.29' 3.23
2 2.85 2.71 2.03 2.18
3 2.60 2.48 209 2.25
4 3.02 2.88 2.14 2.30
5 1.41 1.34 2.15 2.20
6 222 211 2.85 3.06

E-I— 2m — .......
PALMAR DIGITAL A. PROXIMAL TO DORSAL PHALANGEAL A.

HORSE PD IMAGE SIZE (mm) PD VESSEL SIZE (mm) LM IMAGE SIZE (mm) [M VESSEL SIZE (mm)
1 1.75 1.67 3.28' 2.47
2 1.74 1.66 1.73 1.86
3 2.01 1.91 1.82 1.96
4 2.63 2.50 206 2.22
5 1.27 1.21 2.10 2.26
6 2.05 1.95 1.90 2.04

XTSD 1.830.413 21410.22
PALMAR DIGITAL A. ENTERING SOLAR CANAL

HORSE PDIMAGESIZE (mm) PDVI-SSELSIZE (mm) LMIMAGESIZE (mm) LMVESELSIZE (mm)
1 1.80 1.71 2.73’ 2.05
2 1.64 1.56 N / A N / A
3 2.35 2.24 N/ A N / A
4 2.17 2.07 1.43 1.54
5 0.88 0.84 1.81 1.95
6 1.27 1.21 1.81 1.95

 

 

 

 

 

'lnhorse1,thel.Mviewwasreoordedinthe6inchmodeoftl’1eimageintensiﬁermbe. Allotherimageswererecordedin
the9inchmode.

110

111

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 1 1 continued
BULBAR A.
HORSE PD MACE SIZE (mm) PD VESSEL SIZE (mm) LM IMAGE SIZE (mm) LM VE$EL SIZE (mm)
1 N/ A N / A 1.32. 0.99
2 N / A N/ A 0.99 1.06
3 N/ A N / A 2.38 2.55
4 N/ A N/ A 1.35 1.45
5 N / A N / A 1 .06 1.14
6 N/ A N/A 1.83 1.97
3';— m .53....
CORONARY A.
HORSE PD MACE SIZE (mm) PD VE$EL SIZE (mm) LM [MACE SIZE (mm) LM VESEL SIZE (mm)
1 N/ A N / A 1.12. 0.84
2 0.84 0.80 0.68 0.73
3 N/ A N/ A 1.54 1.66
4 N/ A N/ A 0.83 0.89
5 N/ A N/ A 1.05 1.13
6 N / A N/ A 1.21 1.30
stD 0.80 1.0911035
DORSAL PHALANCEAL A.
HORSE PD IMAGE SIZE (mm) PD VE$EL SIZE (mm) LM IMAGE SIZE (mm) [M m SIZE (mm)
1 1.20 1.14 2.11. 1.59
2 1.50 1.43 0.86 0.92
3 1.28 1.22 1.72 1.85
4 0.89 0.85 1.01 1.09
5 1.03 0.98 0.85 0.91
6 1.20 1.58 1.70
x51) 1.1mm _ 1.34.0.9

 

 

 

 

112

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Table 1 1 continued
PAIMAR PHALANGEAL A.

HORSE PDIMAGESIZE (mm) PDVESSELSIZE(mm) LMIMAGESIZE (mm) LMVESSELSIZE(mm)
1 0.95 0.90 1.53' 1.15
2 0.85 0.81 0.78 0.84
3 N/ A N/ A 0.77 0.83
4 0.60 0.57 0.65 0.70
5 0.65 0.62 1.11 1.19
6 0.80 0.76 1.01 _ 1.09

XiSD 07310.14 — 0.973020
DISTAL DORSAL PHALANGEAL A.

HORSE PD IMAGE SIZE (mm) PDVESSELSIZE (mm) LMIMAGESIZE (mm) IMVESSELSIZE (mm)
1 154 1.28 3.96‘ 2.98
2 1.05 1.00 1.09 1.17
3 0.72 0.69 0.97 1.04
4 151 1.44 135 1.45
5 1.43 1.36 1.76 1.89
6 0.91 0.87 1.58 1.70

ﬂ— 1.11.0.0 1.71.0.7.
DISTAL PALMAR PI-IALANGEAL A.

I-IORSE PDIMAGE SIZE (mm) PDVESSELSIZE (mm) LMIMAGESIZE (mm) LMVESSELSIZE (mm)
1 1.16 1.10 4.04' 3.04
2 0.83 0.79 0.85 0.91
3 0.77 0.73 0.96 1.03
4 0.80 0.76 0.89 0.96
5 1.27 1.21 1.09 1.17
6 127 1.21 1.09 1.17

 

 

 

 

 

 

Table 11 continued

113

 

DORSAL BRANCH OF DISTAL DORSAL PHALANGEAL A.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HORSE PDIMACESIZE (mm) PDVE$ELSIZE (mm) LMIMACESIZE (mm) LMVESSELSIZE (mm)
1 1.14 1.09 2.63‘ 1.98
2 0.65 0.62 0.89 0.96
3 0.96 0.91 1.05 1.13
4 0.87 0.83 0.75 0.81
5 0.76 0.72 0.73 0.78
6 1.03 0.98 1.47 1.58
X15D 0.861017 12110.48
PALMAR BRANCH OF DISTAL DORSAL PHALANCAL A.
HORSE PDIMACESIZE (mm) PDVE$ELSIZE (mm) LMIMACESIZE (mm) LMVE$ELSIZE (mm)
1 0.57 0.54 1.89. 1.42
2 0.62 0.59 0.51 0.55
3 0.91 0.87 0.95 1.02
4 0.91 0.87 0.68 0.73
5 0.93 0.89 0.92 0.99
6 0.76 0.72 0.66 0.71
x... — .7... H...
TERMINAL ARCH (AT SACI'IT AL MIDIM)
HORSE PD IMAGE SIZE (mm) PD VE$EL SIZE (mm) [M IMAGE SIZE (mm) LM VESSEL SIZE (mm)
1 1.37 1.30 3.25. 2.44 7
2 1.64 1.56 1.13 1.21
3 1.49 1.42 N/ A N/ A
4 1.88 1.79 1.40 1.51
5 1.16 1.10 2.06 2.22
6 1.42 1.35 1.49 1.60
7..:— ....... ......

 

 

 

 

Table 11 continued

114

 

DORSAL BRANCH OF TERMINAL ARCH

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HORSE PD IMAGE SIZE (mm) PD VESSEL SIZE (mm) LM IMAGE SIZE (mm) LM VESSEL SIZE (mm)
1 0.65 0.62 1.86' 1.40
2 1.00 0.95 0.36 0.39
3 0.91 0.87 0.74 0.80
4 0.53 0.50 0.58 0.62
5 0.76 0.72 0.94 1.01
6 0.64 0.61 0.92 0.99
SOLAR BRANCH OF TERMINAL ARCH
HORSE PD IMAGE SIZE (mm) PD VE$EL SIZE (mm) LM IMAGE SIZE (mm) LM VE$EL SIZE (mm)
1 0.97 0.92 1.54. 1.16
2 0.86 0.82 0.40 0.43
3 0.90 0.86 0.77 0.83
4 1.30 1.24 0.60 0.65
5 0.78 0.74 0.78 0.84
6 0.57 0.54 1.23 1.32
SOLAR MARGIN A.
HORSE PDIMAGESIZE (mm) PDVESSELSIZE (mm) IMIMAGESIZE (mm) IMVE$ELSIZE (mm)
1 0.71 0.68 0.90“ 0.68
2 0.73 0.70 0.68 0.73
3 1.10 1.05 N/ A N/ A
4 0.81 0.77 0.66 0.71
5 0.75 0.71 0.63 0.68
6 0.95 0.90 0.84 0.90
f1- 0.8&t.0.15 07410.09

 

 

 

 

115

Paired T-test for comparative analysis of vessel diameters

in palmarodorsal and lateromedial view angiograms

STATISTIX 3.1
ID: vessel size

PAIRED T TEST FOR PD - LM

MEAN -2.643E-01 STD ERROR 5.972E-02

T -4043
DF 73
P 0.0000

CASES INCLUDED 74 MISSING CASES 16

116

STATISTIX 3.1
ID: vessel size

RANKITS VS PD

RANKITS
3.0 +
+
+
2
+++++
1.0 + 2+2+ +
3233
34+4
75
272
-1.0 + +6
23
++
+
+
-3.0 + '
-+ --------- + --------- + --------- + --------- +-
0.3 1.0 1.7 2.4 3 1

APPROX. WILK-SHAPIRO 0.8700 78 CASES PLOTTED

STATISTIX 3.1
ID: vessel size

RANKITS vs LM

 

RANKITS
3.0 +
+
+
+ 2
3 2
1.0 + 22 4
222 33
4+223+
535
372
-1.0 + 44
5
+++
+
+
-3.0 +
-+ --------- + --------- + --------- + --------- +-
0.2 1 O 1.8 2.6 3 4
LM

APPROX. WILK-SHAPIRO 0.9227 86 CASES PLOTTED

LIST OF REFERENCES

LIST OF REFERENCES

1. Curry TS, Dowdy IE, Murry RC. Christensen's physics of diagnostic
radiology, 4th ed. Philadelphia: [ea 8: Febiger, 1990.

2. General Electric medical systems Advantxe’ specifications manual.
Direction 46-019357, revision 3, 1992.

3. Kruger RA, Reiderer S]. Basic concepts of digital subtraction angiography.
Boston: G. K. Hall Medical Publishers, 1984;27.

4. 3M Service manual, M959 Laser Imager XL, 3M Medical Imaging Systems
Division, 1991.

5. Webster's new collegiate dictionary. Sprinfield: G. 8: C. Merriam
Company, 1974.

6. Greenspan RH. Magnification angiography. In: Abrams angiography-
vascular & interventional radiology, 3rd ed. Abrams, ed. Boston: Little, Brown and
Company, 1983;205-216.

7. Iohnsrude 18, Jackson DC. A practical approach to angiography. Boston:
Little, Brown and Company, 1979.

8. Meridith W], Massey ID. Ch.18 Geometric factors which inﬂuence the
radiographic image. In: Fundamental physics of radiology, 3rd ed. Chicago: John
Wright 8: Sons LTD Publication, 1977 ;211-232.

117

118

9. Eastman Kodak Co. The fundamentals of radiography, 12th ed.
Rochester: Eastman Kodak Co., 1980;23—25.

10. Tortorici MR. Fundamentals of angiography. St. Louis: The CV. Mosby
Company, 1982;3-35.

11. Sugimoto K, Sakurai N, Kaneko M, Shirasawa H, Shibata K, Miyata M,
Noguchi T, Uematsu K, Shimoda K, Sakata 1. Application of renal
microangiography to normal and diseased kidneys of cattle and mice. Am JVet Res
1991;52(1)157-163.

12. General Electric service engineers; personal communications, 1994.

13. Sprawls P. Intensified radiography. In: The physical principles of
diagnostic radiology. Baltimore: University Park Press, 1977 :237-270.

14. Abrams HI... The opaque media: physiologic effects and systemic
reactions. In: Abrams angiography-vascular 8: inter-ventional radiology, 3rd ed.
Abrams, ed. Boston: Little, Brown and Company, 1983;15-39.

15. White RI. Fundamentals of vascular radiology. Philadelphia: Lei 6:
Febiger, 1976;27-42.

16. Katzberg RW, Morris TW, Schulman G, Faillace RT, Boylan LM, Foley
M], Spataro RF, Fischer HW. Reactions to intravascular contrast media, part I: severe
and fatal cardiovascular reactions in a canine dehydration model. Rad 1983;147:327-
330.

17. Katzberg RW, Morris TW, Schulman G, Caldicott W], Boylan LM, Foley
M], Spataro RF, Fischer HW. Reactions to intravascular contrast media, part II: acute
renal response in euvolemic and dehydrated dogs. Rad 1983;147:331-334.

18. Almen T. Development of nonionic contrast media. Invest Rad
1985;20(Supp):82-S9.

119

19. Strain WH, Rogoff SM. The radiopaque media: nomenclature and
chemical formulas. In: Angiography, 2nd ed. Abrams, ed. Boston: Little, Brown and
Company, 1971.

20. Dennis R, Herrtage MB. Low osmolar contrast media, a review. Vet Rad
1989;30(1):2—12.

21. Dawson P. Chemotoxicity of contrast media and clinical adverse effects: a
review. Invest Rad 1985;Supp;584—S91.

22. Bettmann MA, Bourdillon PD, Barry WH, Brush KA, Levin DC. Contrast
agents for cardiac angiography: effects of a nonionic agent vs. a standard ionic agent.
Rad1984;153:583-587.

23. Steinberg EP, Moore RD, Powe NR, Gopalan R, Davidoff AI, Litt M,
Graziano S, Brinker IA. Safety and cost effectiveness of high-osmolality as compared
with low-osmolality contrast material in patients undergoing cardiac angiography.
NEIM 1992;326(7):425-430.

24. Spears IR, Sandor T, Als AV, Malagold M, Markis IE, Grossman W, Serur
IR, Paulin S. Computerized image analysis for quantitative measurement of vessel
diameter from cineangiograms. Circulation 1983;68(2):453-461.

25. Linford RL, O'Brien TR, Wheat ID, Meagher DM. Radiographic
assessment of epiglottic length and pharyngeal and laryngeal diameters in the
thoroughbred. AIVR 1983;44(9):1660-1666.

26. Myer CW. Radiography review: the vascular and bronchial patterns of
pulmonary disease. Vet Rad 1980;21(4):156—160.

27. van den Broek AHM, Darke PGG. Cardiac measurements on thoracic
radiographs of cats. ISmall Anim Prac 1987;28:125-135.

28. Suter PF, Lord PF. Thoracic radiography. Wettswil, Swizterland: Peter F.
Suter, 1984.

 

120

29. Morton WG, Hammer EW. The xray, or, photography of the invisible and
its value in surgery. New York: American Technical Book Co., 1986.

30. Wise M. N on-selective angiocardiography in the normal dog and cat. Vet
Rad 1982;23(4):144-151.

31. Carlsten I. Imaging of the equine heart: an angiographic and
echocardiographic investigation. Uppsala, Merkanﬁl-Tryckeriet AB, 1986.

32. Stickle RL, Anderson LK. Diagnosis of common congenital heart
anomalies in the dog using survey and nonselective contrast radiography. Vet Rad
1987;28(1):6-12.

33. Singh AP, Singh GR, Sharma DN, Nigam IM, Bhargava AK.
Arteriographic anatomy of the abdominal aorta in the goat, dog, pig, and rabbit. Vet
Rad 1982;23(6):279-281.

34. Singh AP, Singh GR, Bhargava AK. Angiographic evaluation of bovine
intestinal healing following inverting, everting, and end-on anastomosis. Vet Rad
1983;24(1):35-40.

35. Singh AP, Nigam IM. Vascular response to fracture healing in the bovine,
an angiographic shrdy. Vet Rad 1983;24(4):174-180.

36. Rendano VT, Georgi IR, White KK, Sack WO, King IM, Bianchi DG,
Theodorides VI. Equine verminous arteritis. an arteriographic evaluation of the
larvicidal activity of albendazole. Eq Vet I. 1979;11(4):223—231.

37. Rantanen NW, Ewing RL. Principles of ultrasound application in animals.
Vet Rad 1981;22(5):196—203.

38. Wallace KD, Selcer BA, Tyler DE, Brown I. Transrectal ultrasonography
of the cranial mesenteric artery of the horse. Am IVet Res 1989;50(10):1699-1703.

 

121

39. Wallace KD, Selcer BA, Becht IL. Technique for trans-rectal
ultrasonography of the cranial mesenteric artery of the horse. Am I Vet Res
1989;50(10):1695—1698.

40. Edwards GB, Allen WE. Aorto-iliac thrombosis in two horses: clinical
course of the disease and use of real-time ultrasonography to confirm diagnosis. Eq
Vet I 1988;20(5):384—386,391.

41. Reef VB, Roby KA, Richardson DW, Vaala WE, Johnston IK. Use of
ultrasonography for the detection of aortic-iliac thrombosis in horses. IAVMA
1987;190(3):286-288.

42. Hedrick WR, Hykes DI.. Doppler physics and insh'umentation, a review.
IDMS 1988;4z109-120.

43. Scoutt LM, Zawin ML, Taylor KI. Doppler US, part II. clinical
applications. Rad 1990;174:3109-3519.

44. Welch RD, Dean PW, Miller MW. Pulsed spectral doppler evaluation of a
peripheral arteriovenous fistula in a horse. IAVMA 1992;200(9):1360-1361.

45. Adair HS, Goble DO, Shires GM, Sanders WL. Evaluation of laser
doppler ﬂowmetry for measuring coronary band and laminar microcirculatory
blood ﬂow in clinically normal horses. Am IVet Res 1994;55(4):445—449.

46. Thomson CE, Kornegay IN, Burn RA, Drayer BP, Hadley DM, Levesque
DC, Gainsburg LA, Lane SB, Sharp NI, Wheeler SI. Magnetic resonance imaging-a
general overview of principles and examples in veterinary neurodiagnosis. Vet Rad
8: US 1993;34(1):2-17.

47. Edelman RR. MRI angiography: approaches and strategies. MRI decisions
1989:13-13—23.

48. Koblik PD, Homof WI, Rhode EA, Kelly AB. Left ventricular ejection
fraction in the normal horse determined by ﬁrst-pass nuclear angiocardiography.
Vet Rad 1985;26(2):53-62.

122

49. Koblik PD, Hornof WI. Diagnostic radiology and nuclear cardiology:
their use in assessment of equine cardiovascular disease. In: VCNA, Eq Prac, I
Bonagura (guest ed.) Philadelphia: W.B. Saunders Company, 1985; 289-309.

50. Daniel GB, Wantschek L, Bright R, Silva-Krott 1. Diagnosis of aorh'c
thromboembolism in two dogs with radionuclide angiography. Vet Rad
1990;31(4):182-185.

51. Gogoi SN, Nigam IM, Singh AP. Angiographic evaluation of bovine foot
abnormalities. Vet Rad 1982;23(4):171-174.

52. Burns I, Cornell C. Angiography of the caprine digit. Vet Rad
1981;22(4):174-176.

53. Coffman IR, Iohnson IH, Guffy MM, Finocchio EI. Hoof circulation in
equine laminitis. IAVMA 1970;156(1):76-83.

54. Ackerman N, Garner HE, Coffman IR, Clement IW. Angiographic
appearance of the normal equine foot and alterations in chronic laminitis. IAVMA
1975;166(1):58-62.

55. Scott EA, Thrall DE, Sandler GA. Angiography of equine metacarpus and
phalanges: alterations with medial palmar artery and medial palmar digital artery
ligation. Am IVet Res 1976;37(8):869-873.

56. Scott EA, Sandler GA, Shires MH. Angiography as a diagnostic technique
in the equine. I Eq Med Surg 1978;2(6):270-278.

57. Lindbom A. Arterial spasm caused by puncture and catheterization. Acta
radiol 1957;449-459.

58. van Kraayenburg F], Fairall N, Littlejohn A. The effects of vertical force
on blood ﬂow in the palmar digital arteries of the horse. Proc lst Int Conf, Equine
Exercise Physiology, Oxford, 1982; 144-154.

123

59. Elmes PI, Eyre P. Vascular reactivity of the bovine foot to neurohormones,
antigens, and chemical mediators of anaphylaxis. Am IVet Res 1977;38(1):107-112.

60. Harkema IR, Robinson NE, Scott IB. Cardiovascular, acid-base,
electrolyte, and plasma volume changes in ponies developing alimentary laminitis.
Am IVet Res 1978;39(5):741-744.

61. Schummer A, Wilkens H, Vollmerhaus B, Habermehl KH. The circulatory
system, the skin, and the cutaneous organs of the domestic mammals. New
York:Springer Verlag, 1981;552-597.

62. Colles CM, Garner HE, Coffman IR. The blood supply of the horse's foot.
Proc25thAnthg,AmAssocEqPrac, Miami, FL, 1980;385-389.

63. Mishra PC, Leach DH. Extrinsic and intrinsic veins of the equine hoof
wall. I Anat 1983;136(3):543-560.

64. Stashak TS. Adams' lameness in horses. 4th ed. Philadelphia: Lea 8:
Febiger, 1987;14.

65. Pepine CI. Diagnostic and therapeutic cardiac catheterization. 2nd edition.
Philadelphia: Williams and Wilkins, 1994; 171.

 

"Ililljlll01117111“