Morphology of distal border synovial invaginations of the equine distal sesamoid bone: comparison between computed tomography and a hoof-specific radiographic projection


Claerhoudt, S.; Bergman, H.J.; van der Veen, H.; Duchateau, L.; Raes, E.V.; Vanderperren, K.; Saunders, J.H.

Veterinary and Comparative Orthopaedics and Traumatology 25(6): 453-459

2013


To compare the difference and agreement of the morphology of distal border synovial invaginations on a dorsoproximal-palmarodistal oblique (DPr-PaDiO) projection with hoof-specific angle versus computed tomography. Computed tomography (CT) images and a DPr-PaDiO radiographic projection with hoof-specific angle were obtained on 50 cadaveric forefeet from 25 Warmblood horses. Computed tomography was assumed to be the gold standard. The number, shape and depth of penetration of distal border synovial invaginations into the distal sesamoid bone were evaluated with both methods, and the comparison of their measurements was statistically described. Significantly more invaginations were seen on CT compared to radiography, with an observed average difference of 1.2. In none of the cases did radiography have a higher number than that observed with CT. No statistically significant difference for depth between CT and the DPr-PaDiO projection was seen, however, there was quite a large variation of the actual difference of measurements against their mean found. Radiography was underestimated when high mean values applied. The agreement between both modalities for shape was moderate to good. A very high specificity of the specific DPr-PaDiO projection for shape was found (97%). The radiographic projection with hoof-specific angle differs significantly from CT concerning the number and depth of the distal border synovial invaginations. Therefore, this specific view may not be considered useful in the evaluation of these invaginations.

Original
Research
Schattauer
2012
453
Morphology
of
distal
border
synovial
invaginations
of
the
equine
distal
sesamoid
bone
Comparison
between
computed
tomography
and
a
hoof-specific
radiographic
projection
S.
Claerhoudt';
H.
J.
Bergmann;
H.
van
der
Veen
2
;
L.
Duchateau
3
;
E.
V.
Raes';
K.
Vanderperren';
J.
H.
Saunders'
'Veterinary
Medical
Imaging
and
Small
Animal
Orthopaedics,
Faculty
of
Veterinary
Medicine,
Ghent
University,
Merelbeke,
Belgium;
2
Lingehoeve
DiergeneeskundeNetCT,
AK
Lienden,
Netherlands;
3
Department
of
Physiology
and
Biometry,
Faculty
of
Veterinary
Medicine,
Ghent
University,
Merelbeke,
Belgium
Keywords
Distal
sesamoid
bone,
warmblood
horse,
synovial
invagination,
radiography,
computed
tomography
Objectives:
To
compare
the
difference
and
agreement
of
the
morphology
of
distal
border
synovial
invaginations
on
a
dorsoproximal-
palmarodistal
oblique
(DPr-PaDiO)
projection
with
hoof-specific
angle
versus
computed
to-
mography.
Methods:
Computed
tomography
(CT)
im-
ages
and
a
DPr-PaDiO
radiographic
projection
with
hoof-specific
angle
were
obtained
on
50
cadaveric
forefeet
from
25
Warmblood
horses.
Computed
tomography
was
assumed
to
be
the
gold
standard.
The
number,
shape
and
depth
of
penetration
of
distal
border
syn-
ovial
invaginations
into
the
distal
sesamoid
bone
were
evaluated
with
both
methods,
and
the
comparison
of
their
measurements
was
statistically
described.
Correspondence
to:
Sarah
Claerhoudt
Faculty
of
Veterinary
Medicine
Department
of
Veterinary
Medical
Imaging
and
Small
Animal
Orthopaedics
Salisburylaan
133
Merelbeke
9820
Oost-Vlaanderen
Belgium
Phone:
+
32
92
647650
Fax:
+
32
92
64
77
93
E-mail:
Results:
Significantly
more
invaginations
were
seen
on
CT
compared
to
radiography,
with
an
observed
average
difference
of
1.2.
In
none
of
the
cases
did
radiography
have
a
higher
number
than
that
observed
with
CT.
No
statistically
significant
difference
for
depth
between
CT
and
the
DPr-PaDiO
projection
was
seen,
however,
there
was
quite
a
large
variation
of
the
actual
difference
of
measure-
ments
against
their
mean
found.
Radiography
was
underestimated
when
high
mean
values
applied.
The
agreement
between
both
moda-
lities
for
shape
was
moderate
to
good.
A
very
high
specificity
of
the
specific
DPr-PaDiO
pro-
jection
for
shape
was
found
(97%).
Clinical
significance:
The
radiographic
pro-
jection
with
hoof-specific
angle
differs
signifi-
cantly
from
CT
concerning
the
number
and
depth
of
the
distal
border
synovial
invagi-
nations.
Therefore,
this
specific
view
may
not
be
considered
useful
in
the
evaluation
of
these
invaginations.
Vet
Comp
Orthop
Traumatol
2012;
25:
453-459
doi:10.3415NCOT-11-10-0149
Received:
October
21,
2011
Accepted:
June
12,
2012
Pre-published
online:
July
27,
2012
Introduction
Although
'abnormal'
distal
border
synovial
invaginations
of
the
distal
sesamoid
bone
are
described
as
being
related
with
navicu-
lar
disease,
their
clinical
relevance
still
re-
mains
unclear.
The
presence
of
more
than
seven
synovial
invaginations,
inverted
flask-shaped,
or
moderate
to
deeply
pen-
etrating
invaginations
along
the
horizontal
or
sloping
distal
navicular
borders
are
con-
sidered
a
poor
condition
and
more
likely
of
clinical
significance
(1-10).
During
purchase
examinations,
the
syn-
ovial
invaginations
are
often
graded
ac-
cording
to
a
radiographic
classification
sys-
tems
using
a
dorso
45°-70°
proximal-pal-
marodistal
oblique
(D45°-70°Pr-PaDiO)
radiographic
projection.
A
recent
study
has
reported
poor
agreement
concerning
the
morphology
of
the
synovial
invaginations
between
three
radiographic
projections
(D45°Pr-PaDiO,
D55°Pr-PaDiO
and
D65°Pr-PaDiO)
and
computed
to-
mography
(CT),
which
is
used
as
the
gold
standard
(11).
Three
factors
were
suggested
to
explain
the
poor
agreement:
the
variable
height
of
the
heels,
the
different
orien-
tations
of
synovial
invaginations
into
the
distal
sesamoid
bone,
and
the
superimposi-
tion
of
the
distal
sesamoid
bone
over
other
structures
on
a
DPr-PaDiO
projection
(12).
Consequently,
the
use
of
radiography
for
evaluation
of
the
synovial
invaginations
in
purchase
examinations
was
questioned.
In
the
present
study,
the
authors
hypo-
thesized
that
it
should
be
possible
to
select
a
DPr-PaDiO
projection
that
allowed
opti-
mal
evaluation
of
the
majority
of
synovial
Vet
Comp
Orthop
Traumatol
6/2012
454
S.
Claerhoudt
et
al.:
Morphology
of
synovial
invaginations
of
the
equine
distal
sesamoid
bone
invaginations
in
the
individual
horse.
For
this,
the
degree
of
angulation
of
the
DPr-
PaDiO
projection
is
selected
based
on
the
height
of
the
heels
and
the
most
common
`straight'
direction
of
the
invaginations
in
the
distal
sesamoid
bone
(12).
The
aim
of
the
study
was
to
compare
the
morphology
of
the
synovial
invaginations
on
this
hoof-
specific
DPr-PaDiO
projection
with
CT.
Material
and
methods
Material
The
material
used
in
this
study
consisted
of
50
forefeet
of
25
Warmblood
horses
(mean
age:
7
years).
All
horses
were
subjected
to
euthanasia
for
reasons
unrelated
to
this
study.
All
feet
were
severed
at
the
level
of
the
metacarpophalangeal
joint
immediately
after
euthanasia.
The
shoe
and
loose
horn
in
the
sole,
if
present,
were
removed
and
the
frog
was
cleaned.
The
feet
were
not
selected
by
any
particular
criteria
and
both
forefeet
of
each
horse
were
included.
Computed
tomography
examination
The
CT
scans
were
performed
with
a
four-
detector
row
spiral
CT
scanners
in
which
the
feet
were
placed
in
the
gantry
with
the
longitudinal
axis
of
the
foot
oriented
paral-
lel
to
the
CT
table
and
perpendicular
to
the
plane
of
the
CT
gantry.
The
medial
side
of
the
foot
was
marked.
The
limbs
were
scanned
in
a
distal
to
proximal
direction.
The
output
parameters
were
120
kV
and
250
mAs
per
slice.
The
slice
thickness
was
0.6
mm,
pitch
of
0.875
cm,
0.3
increment
and
1-second
rotation
time.
Transverse
CT
scans
were
reconstructed
from
the
level
of
the
distal
to
the
proximal
aspect
of
the
dis-
tal
sesamoid
bone
using
a
bone
window
setting
(WL:
200-600;
WW:
1000-2000),
250
mm
field
of
view
and
512
x
512
pixel
matrix.
The
average
total
time
required
for
scanning
of
each
foot
was
46.25
seconds.
From
the
transverse
images,
dorsal
recon-
e
Mx8000:
Philips
Medical
Systems,
AE
Eindhoven,
The
Netherlands
Fig.
1
Illustration
of
the
specific
slope
measurement
using
the
lateromedial
radio-
graphic
projection.
structions
with
a
slice
thickness
of
0.6
mm
were
reformatted
by
use
of
softwareb.
Radiographic
examination
Radiographic
examination
was
performed
after
the
CT
examination.
The
sulci
of
the
frog
were
packed
with
modelling
com-
poundc.
A
lateromedial
(LM)
and
DPr-
PaDiO
radiographic
projection
were
per-
formed
on
all
feet.
The
LM
projection
was
made
without
a
grid
with
the
foot
placed
on
a
flat
wooden
block,
to
calculate
the
spe-
cific
slope
used
for
the
DPr-PaDiO
projec-
tion.
On
the
LM
projection,
first
a
line
was
drawn
parallel
to
the
flexor
cortex
of
the
distal
sesamoid
bone.
A
second
line
was
drawn
through
the
most
proximal
point
of
the
extensor
process
of
the
distal
phalanx
perpendicular
to
the
first
line.
The
angle
between
this
latter
line
and
a
horizontal
line,
representing
the
flat
wooden
block,
was
measured
(
Fig.
1).
Further,
the
foot
was
placed
on
a
wooden
block
with
a
slope
of
45°
with
the
horizontal,
and
by
the
use
of
wedges
(slope
of
2°,
3°,
and
with
the
horizontal),
a
DPr-PaDiO
projection
with
specific
slope
for
a
particular
foot
was
made,
using
a
grid.
If
the
measured
angle
was
lower
than
45°,
a
construction
was
made
and
the
feet
were
angled
until
the
specific
angle
was
reached.
The
radio-
graphic
beam
of
the
tubed
was
kept
hori-
b
Osirix
Image
processing
Software:
OsiriX,
Los
Angeles,
California,
USA
Playdoh®:
Rainbow
Crafts,
Cincinnati,
Ohio,
USA
d
Mobilux,X-rayEquipment
Verachtert,
Antwerpen,
Belgium
zontal
and
centered
2
cm
proximal
to
the
coronary
band
at
the
midline
of
the
foot.
By
using
this
hoof-specific
angle,
the
beam
was
directed
perpendicular
to
the
flexor
cortex
of
the
distal
sesamoid
bone,
thus
perpendicular
to
the
straight
running
in-
vaginations.
The
feet
were
radiographed
using
60
kV
and
12.5
mAs,
a
grid
(6:1
ratio,
103
lines/cm)
and
a
100
cm
focus-film
dis-
tance.
Image
analysis
Two
observers,
one
board-certified
radiol-
ogist
(JHS)
and
one
PhD-student
(SC),
in-
terpreted
all
CT
and
radiographic
images
together
and
a
diagnosis
was
made
in
con-
sensus.
The
radiographic
image
of
a
par-
ticular
foot
and
hoof
angle
were
reviewed
in
a
randomized
order
at
the
same
work-
station,
on
the
same
diagnostic
imaging
screense
and
using
a
similar
evaluation
scheme
to
determine
the
number
of
synov-
ial
invaginations.
Furthermore,
for
each
synovial
invagination,
the
depth
and
shape
were
determined.
Next,
the
CT
images
of
a
particular
foot
were
reviewed
in
a
random
order
as
well
to
determine
the
number
of
synovial
invaginations
using
transverse
slices
and
dorsal
reconstructions,
and
for
each
synovial
invagination,
the
depth
and
shape
were
determined.
In
a
second
step,
the
corresponding
radiographic
and
CT
images
were
con-
sidered
together.
To
compare
depth
and
shape
assessments
on
the
two
imaging
mo-
dalities,
only
synovial
invaginations
for
which
an
assessment
was
available
on
both
radiography
and
CT
were
used
(some
in-
vaginations
seen
with
CT
were
not
seen
with
radiography).
The
depth
of
penetration
of
the
synovial
invaginations
was
assessed
on
the
dorsal
CT
and
radiographic
images.
Each
synovial
invagination
was
calculated
by
imaging
softwareb
using
the
following
equation:
Depth
(R)
=
A/B,
where
A
=
distance
(in
centimetres)
between
the
most
distal
basis
and
the
proximal
top
of
the
synovial
invagi-
nations,
B
=
distance
(in
centimetres)
be-
tween
the
distal
and
proximal
flexor
e
Totoku
monochrome
LCD
display:
Lewisville,
USA
Vet
Comp
Orthop
Traumatol
6/2012
0
Schattauer
2012
A
C
liwir".F.'c
qiiita...
d
iae
l
6
mo
B
Fig.
2
Dorsally
reconstructed
computed
tomographic
(A)
and
D50°Pr-PaDiO
radiographic
images
(B)
of
a
distal
sesamoid
bone.The
two
most
medially
located
synovial
invaginations
on
the
CT
image
(circle)
are
undetectable
on
the
radiographic
image.
Lateral
is
on
the
left
side
and
medial
on
the
right
side
of
the
image.
S.
Claerhoudt
et
al.:
Morphology
of
synovial
invaginations
of
the
equine
distal
sesamoid
bone
455
borders
of
the
distal
sesamoid
bone.
Data
of
depth
were
classified
in
three
categories:
1
=
R
a1.33,
2
=
0.33<ll
g).5
and
3
=
R
>0.5.
The
shape
of
the
synovial
invaginations
was
assessed
on
dorsal
CT
and
radio-
graphic
images.
The
shape
could
be
cat-
egorized
into
four
categories:
conical,
linear,
lollipop
or
branched.
For
statistical
aims,
this
group
of
four
shapes
will
be
further
described
as
`shape4.
Additionally,
this
group
was
divided
into
two
groups
(further
described
as
`shape2'),
with
group
1
representing
the
normal
(conical-
or
lin-
ear-shaped),
and
group
2
the
abnormal
shapes
(lollipop-
or
branched-shaped).
Since
it
is
described
that
the
synovial
in-
vaginations
run
in
different
directions
in
the
distal
sesamoid
bone,
the
synovial
in-
vaginations
were
classified
as
'straight'
or
Vorsoproximal/
palmaroproximal'
oriented
invaginations
(12).
Statistical
analysis
To
compare
the
observed
number
and
depth
of
the
synovial
invaginations
be-
tween
CT
and
the
DPr-PaDiO
projection,
a
paired
t-test
was
used
with
foot
as
block
variable
for
the
number
and
invagination
as
block
factor
for
the
depth.
The
results
were
summarized
by
the
average
difference
and
corresponding
95%
confidence
inter-
val
and
95%
reference
interval.
The
95%
reference
interval
is
given
by
the
mean
dif-
ference
±
2
times
the
standard
deviation,
which
contains
95%
of
the
actual
differ-
ences
if
the
normal
distribution
assump-
tion
holds.
Bland
Altman
plots
are
pro-
vided
for
CT
versus
the
DPr-PaDiO
projec-
tion,
to
investigate
a
possible
relationship
between
the
difference
and
the
magnitude
of
the
measurement.
The
effect
of
the
type
of
orientation
of
the
synovial
invaginations
on
the
difference
between
CT
and
radiography
for
number
and
depth,
was
evaluated
by
a
mixed
model
with
horse
and
horse
nested
in
foot
as
ran-
dom
effects.
All
tests
were
based
on
a
sig-
nificance
level
of
five
percent
and
perform-
ed
using
statistical
software
f
.
The
degree
of
agreement
between
CT
and
the
DPr-PaDiO
projection
for
number
and
f
SAS
version
9.2.:
SAS
Corporation,
Cary,
NC,
USA
depth
(taken
as
categorical
variables)
was
quantified
using
the
weighted
and
unweight-
ed
(or
simply
'lc')
kappa
statistic.
The
degree
of
agreement
for
shape
was
quantified
using
the
unweighted
kappa
statistic.
The
guide-
lines
for
strength
of
agreement
based
on
the
values
of
lc
were:
<0.20
poor,
0.21-0.40
fair,
0.41-0.60
moderate,
0.61-0.80
good
and
0.81-1.00
very
good
(17).
Results
The
foot-specific
angle
ranged
from
40°
to
62°,
with
an
average
angle
of
51.3°
±
5.2.
Three
feet
(6%)
had
a
measured
angle
lower
than
45°.
Number
The
average
number
of
synovial
invagi-
nations
was
5.9
±
1.56
on
CT
(total
of
295)
and
4.08
±
1.61
on
the
DPr-PaDiO
projec-
tion
(total
of
204),
with
116/204
(56.9%)
running
straight
and
88/204
(43.1%)
run-
ning
in
a
dorso-/palmaroproximal
direc-
tion
in
the
distal
sesamoid
bone,
according
to
the
CT
images.
In
only
eight
of
50
(16%)
feet,
the
number
of
synovial
invaginations
counted
on
the
CT
scans
and
on
the
DPr-
PaDiO
projections,
were
equal.
Even
in
three
of
the
50
(6%)
feet,
there
were
not
any
synovial
invaginations
detected
on
the
radiographic
projection,
however
three,
four
and
six
invaginations,
respectively,
were
counted
on
the
CT
images.
In
none
of
the
cases
did
radiography
have
a
higher
number
observed
than
for
CT
(
Fig.
2).
Significantly
more
synovial
invaginations
were
seen
on
CT
compared
to
radiography
(p
<0.001),
with
an
observed
average
differ-
ence
of
1.2
(95%
CI:
0.86
1.46).
According
to
the
95%
reference
interval
(95%
RI:
—0.99
3.3),
a
large
variation
of
the
actual
differences
was
observed.
The
mean
differ-
ence
for
the
number
of
straight
running
synovial
invaginations
on
CT
versus
radi-
ography
was
equal
to
0.94
(95%
CI:
0.67
1.20)
and
differed
significantly
(p
<0.001)
from
that
of
the
dorso-/palmaroproximal
running
ones,
which
was
equal
to
0.36
(95%
CI:
0.16
0.53).
The
Bland
Altman
plots
of
the
difference
for
number
between
both
modalities
for
all
data
against
their
mean,
is
given
in
Figure
3.
The
weighted-
and
(unweighted)
lc-values
between
the
CT
and
the
particular
radiographic
projection
for
number
of
all
invaginations
was
low
(0.37
and
0.17,
re-
spectively),
representing
a
poor
agreement
between
both
modalities.
C
Schattauer
2012
Vet
Comp
Orthop
Traumatol
6/2012
1
1
1
1
1 1
1
1
2
3
1
1
3
2
2
1
1
1
3
1
4
t
8
3
7
2
4
6
Mean
of
two
readings
-
6
4
.
-
2
I
0
-2
.
4
-4
;
51
-6
8,
0.6-
1
0.4-
:
0.2-
'
-!
0
-0.2-
is
-0.6
1
0.0
00
0
0
0
0.1
0.2
0.3
0.4
Mean
of
two
readings
0
O
00
0
o 0
o
0 0 0
.W
-
0.5
0.6
Fig.
3
Bland
Altman
plots:
Differences
of
counted
numbers
(readings)
on
computed
tomographic
and
radiographic
images
with
the
hoof-specific
angle
against
their
mean.
The
numerical
code
repre-
sents
the
number
of
distal
sesamoid
bones
(total
sum
of
50)
with
the
same
difference
in
number
against
mean.
The
dashed
line
represents
zero-level
(no
difference),
upper
line
=
mean
+
25D,
middle
line
=
mean,
lower
line
=
mean
-
2SD.
SD
=
standard
deviation.
Fig.
4
Bland
Altman
plots:
Differences
of
equations
of
depth
(readings)
on
computed
tomographic
and
radiographic
images
with
the
hoof-specific
angle
against
their
mean.
Each
individual
plot
(0)
rep-
resents
a
synovial
invagination
(total
of
295
plots).The
dashed
line
represents
zero-level
(no
difference),
upper
line
=
mean
+
25D,
middle
line
=
mean,
lower
line
=
mean
-
25D.
SD
=
standard
deviation.
456
S.
Claerhoudt
et
al.:
Morphology
of
synovial
invaginations
of
the
equine
distal
sesamoid
bone
There
was
not
any
statistically
significant
difference
for
depth
between
CT
and
the
DPr-PaDiO
projection
seen
(p
>0.05).
The
mean
difference
between
CT
and
radiogra-
phy
was
equal
to
-0.0001
(95%
CI:
-0.02
-
0.02),
meaning
that
there
seems
to
be
little
or
no
bias.
The
95%
reference
interval
for
the
radiographic
projection
corresponds
to
-
0.22
-
0.22,
meaning
that
quite
a
large
variation
of
actual
differences
between
CT
and
radiography
is
present.
The
mean
dif-
ference
for
the
depth
of
the
straight
run-
ning
synovial
invaginations
measured
on
CT
versus
radiography
was
equal
to
0.01
(95%
CI:
-0.01
-
0.04),
and
-0.02
(95%
CI:
-
0.04
-
0.002)
for
the
dorso-/palmaro-
proximal
running
ones.
None
of
the
two
differed
significantly
from
zero
(p
>0.05).
The
mean
difference
between
the
straight
and
dorso-/palmaroproximal
running
in-
vaginations
was
equal
to
0.03,
which
was
also
not
statistically
significant.
A
Bland
Altman
plot
of
the
differences
for
depth
be-
tween
both
modalities
for
all
data
against
their
mean,
is
given
in
Figure
4.
As
the
average
of
the
CT
and
radiography
measurement
increases
(deeper
synovial
invaginations),
the
difference
also
seem
to
increase,
meaning
that
an
underestimation
of
radiography
was
seen
with
high
mean
values.
The
weighted
and
(unweighted)
lc-valu-
es
between
CT
and
the
DPr-PaDiO
projec-
tion
for
depth
of
all
data
were
equal
to
0.45
and
0.43
respectively,
representing
a
mod-
erate
agreement
between
both
modalities.
The
agreement
between
CT
and
the
radio-
graphic
projection
for
depth
of straight
and
dorso-/palmaroproximal
running
synovial
invaginations
was
comparable
(x-values
of
0.42
and
0.44,
respectively).
Shape
In
three
of
eight
(37.5%)
feet
with
an
equal
number
of
synovial
invaginations
on
both
the
CT
scans
and
the
particular
DPr-PaDiO
projection
respectively,
the
shape
was
com-
parable
on
both
methods.
The
agreement
between
both
modalities
for
shape4
was
moderate
(unweighted
lc-value
of
0.53)
compared
to
shape2,
which
lc-value
was
higher
(0.75),
representing
a
good
agree-
ment.
The
sensitivity
and
specificity
(both
measured
for
shape2)
and
corresponding
95%
CI
of
the
DPr-PaDiO
projection
for
shape,
with
CT
as
gold
standard,
are
sum-
marized
in
Table
1.
Discussion
The
present
study
was
carried
out
to
assess
the
agreement
between
the
morphology
of
distal
border
synovial
invaginations
of
the
distal
sesamoid
bone
on
a
DPr-PaDiO
radiographic
projection
with
hoof-specific
angle
and
CT
scans.
Histology
is
regarded
as
the
gold
standard
for
the
diagnosis
of
tis-
sue
abnormalities
(13).
In
the
present
study,
no
histological
examination
was
per-
formed,
however,
due
to
the
possibility
of
reconstruction
of
multiplanar,
high-resol-
ution
images
without
superimposition,
CT
provides
detailed
anatomical
information
of
the
synovial
invaginations
(14).
There-
fore,
CT
was
assumed
by
the
authors
to
be
the
gold
standard.
The
mean
difference
between
data
on
CT
images
and
the
hoof-specific
radio-
graphic
projections
for
number
was
posi-
tive,
meaning
that
the
visibility
of
the
syn-
ovial
invaginations
was
better
on
CT
com-
pared
to
radiography.
A
similar
conclusion
was
reached
in
previous
studies
(11,
14-16).
In
our
study,
the
mean
difference
between
both
the
modalities
for
number
of
all
data
was
lower
compared
to
the
results
of
a
previous
study
of
the
same
population
(1.2
versus
2)
(11).
In
both
studies,
a
sig-
nificant
mean
difference
between
CT
and
radiography
for
number
of
synovial
invagi-
nations
was
found.
The
mean
difference
for
all
feet
between
CT
images
and
the
hoof-specific
projec-
tions
for
number
of
the
dorso-/palmaro-
Vet
Comp
Orthop
Traumatol
6/2012
C
Schattauer
2012
Sensitivity
Specificity
(95%
CI)
95%
CI)
Total
0.75
(0.60
-
0.87)
0.97
(0.92
-
0.99)
Straight
0.64
(0.43
-
0.82)
0.95
(0.89
-
0.99)
Dorso-/palmaroproximal
0.89
(0.67
-
0.99)
0.98
(0.91
-
1.00)
Sensitivity
and
specificity
of
the
DPr-PaDiO
radiographic
projection
with
hoof-specific
angle
for
shape
of
total,
straight
and
dorso-/palmaroproximal
running
invaginations,
with
computed
tomography
as
gold
standard.
Key:
CI
=
confidence
interval.
A
I
ril
11,
)4
1
V"
Fig.
5
Dorsally
reconstructed
computed
tomographic
(CT)
image
(A)
and
corresponding
D62°Pr-
PaDiO
radiographic
image
(B)
of
a
distal
sesamoid
bone,
showing
a
deeply
penetrating
invagination
on
the
CT
image
(circle),
which
is
mildly
penetrating
on
the
radiographic
image
(circle).
Lateral
is
on
the
left
side
and
medial
on
the
right
side
of
the
image.
S.
Claerhoudt
et
al.:
Morphology
of
synovial
invaginations
of
the
equine
distal
sesamoid
bone
457
proximal
running
synovial
invaginations
was
lower
than
for
the
straight
running
ones
(0.36
versus
0.94),
meaning
that
the
hoof-specific
radiographic
projections
ap-
peared
better
in
the
evaluation
of
the
number
of
dorso-/palmaroproximal
run-
ning
synovial
invaginations.
Unfor-
tunately,
the
first
group
of
invaginations
represents
only
36.6%
of
the
total
of
synov-
ial
invaginations
(compared
to
63.4%
for
the
latter
group)
(12).
Indeed,
the
lc-value
between
CT
images
and
the
DPr-PaDiO
projections
for
number
of
dorso-/palma-
roproximal
running
synovial
invaginations
was
higher
than
for
the
straight
running
ones.
However,
the
mean
difference
be-
tween
CT
images
and
the
hoof-specific
projections
for
number
of
both
types
of
orientation
of
the
synovial
invaginations
in
the
distal
sesamoid
bone,
was
significant.
The
presence
of
these
significant
differ-
ences
between
both
the
modalities
may
be
explained
by
the
lack
of
superimposition
of
surrounding
bone
and
the
high
sensitivity
of
CT
for
detecting
bone
in
detail,
com-
pared
to
radiography.
The
synovial
invagi-
nations
that
were
ill-defined
or
undetect-
able
on
radiography
were
mostly
small
and
located
at
the
distal
sloping
borders
of
the
distal
sesamoid
bone
on
the
CT
images.
The
clinical
significance
of
these
subtle
CT
findings
remains
questionable.
It
has
been
described
that
the
presence
of
small
synov-
ial
invaginations
at
the
sloping
borders
on
CT
may
be
observed
in
normal
distal
sesa-
moid
bones,
but
is
considered
significant
if
seen
on
radiographs
(3,
14).
Results
of
our
study
show
that
the
mean
difference
for
depth
of
the
synovial
invagi-
nations
between
CT
and
the
hoof-specific
radiographic
projection
was
not
signifi-
cant.
The
mean
differences
of
measure-
ments
for
depth
between
CT
and
radiogra-
phy
were
all
close
to
zero,
meaning
that
the
hoof-specific
angle
may
appear
good
for
the
evaluation
of
the
depth.
Nevertheless,
the
reference
intervals
were
all
quite
wide,
representing
a
large
variation
of
the
actual
differences.
This
may
be
explained
by
the
fact
that
most
deep
invaginations
on
CT
were
ending
proximally
as
tiny,
deeply
pen-
etrating
lines,
which
were
undetectable
on
radiography
(
Fig.
5).
We
found
that
when
the
mean
measurement
on
CT
and
radiography
increased,
the
difference
be-
tween
the
measurements
also
seemed
to
in-
crease,
resulting
in
an
underestimation
of
radiography
when
high
mean
values
apply.
Similar
findings
were
seen
in
a
previous
study
(11).
Such
a
result
may
have
clinical
importance
since
reports
in
the
literature
state
that
an
increase
in
the
depth
of
pen-
etration
of
distal
border
synovial
invagi-
nations
is
considered
a
significant
radio-
graphic
finding
(9,
10).
In
the
current
study,
the
sensitivity
and
specificity
of
the
DPr-PaDiO
projection
with
hoof-specific
angle
for
shape
were
high.
A
97%
specificity
and
75%
sensitivity
were
found,
meaning
that
almost
no
false
positive,
but
more
false
negative
results
were
seen.
To
summarize,
if
an
abnormal
shaped
synovial
invagination
is
seen
on
the
hoof-specific
radiographic
projection,
there
is
high
probability
that
it
is
effectively
abnormal,
but
if
a
normal
one
is
seen,
it
can
be
abnormal
on
CT
(
Fig.
6).
The
sensitiv-
ity
and
specificity
for
the
specific
radio-
graphic
projection
were
highest
for
the
dorso-/palmaroproximal
running
synovial
invaginations.
C
Schattauer
2012
Vet
Comp
Orthop
Traumatol
6/2012
It
it
-
NO
I
Air
B
re;
Fig.
6
Dorsally
reconstructed
computed
tomographic
(CT)
image
(A)
and
corresponding
D50°Pr-
PaDiO
radiographic
image
(e)
of
a
distal
sesamoid
bone,
showing
a
branched-shaped
invagination
on
the
CT
image
(circle),
which
is
conical-shaped
on
the
radiographic
image.
Lateral
is
on
the
left
side
and
medial
on
the
right
side
of
the
image.
458
S.
Claerhoudt
et
al.:
Morphology
of
synovial
invaginations
of
the
equine
distal
sesamoid
bone
The
degree
of
agreement
between
measurements
on
CT
images
and
the
hoof-
specific
radiographic
projections
was
highest
for
shape,
compared
to
both
other
variables.
Statistically,
the
ic-value
was
measured
using
a
2x2
table
for
shape2
(2
groups/categories,
see
above)
and
a
4x4
table
for
shape4
(4
cat-
egories).
For
the
resulting
2x2
table,
a
high
un-
weighted
ic-value
(0.75)
was
found,
com-
pared
to
the
4x4
table
(0.53).
In
contrast,
data
of
number
were
classified
in
12
categories
(the
counted
number
of
invaginations
per
bone
varied
from
0
to
maximum
11
invaginations,
equal
to
12
categories),
logically
resulting
in
a
very
low
ic-value
(0.17).
In
theory,
any
value
of
K
much
below
0.5
will
indicate
poor
agree-
ment.
However,
despite
these
published
guidelines,
no
value
of
kappa
can
be
regarded
universally
as
indicating
some
degree
of
agreement
(17,
18).
Actually,
the
value
of
ic
de-
pends
on
the
number
of
categories
and
upon
circumstances,
as
demonstrated
in
our
re-
sults.
In
a
previous
study
of
the
same
authors,
a
poor
agreement
of
the
morphology
of
the
synovial
invaginations
between
the
three
radiographic
projections
and
the
gold
stan-
dard
CT
was
found
(11).
Three
factors
were
suggested
to
explain
these
results:
the
vari-
able
height
of
the
individual
heels,
the
differ-
ent
orientations
of
synovial
invaginations
into
the
distal
sesamoid
bone
and
the
super-
imposition
of
the
distal
sesamoid
bone
over
other
structures
on
a
DPr-PaDiO
projection
(12).
In
the
present
study,
the
authors
hypo-
thesized
that
ruling
out
two
of
the
three
in-
fluencing
factors
by
measuring
an
optimal
angle
based
on
the
height
of
the
heels
and
the
most
common
'straight'
direction
of
the
in-
vaginations
in
the
distal
sesamoid
bone,
a
better
evaluation
of
the
majority
of
synovial
invaginations
in
the
individual
horse
would
be
possible
(12).
In
theory,
when
the
radio-
graphic
beam
is
directed
perpendicular
to
the
straight
running
synovial
invaginations
into
the
distal
sesamoid
bone,
these
invagi-
nations
do
not
shorten
or
enlarge
in
depth,
which
was
the
case
for
the
dorso-
and
palma-
roproximal
oriented
synovial
invaginations,
respectively.
In
our
study,
a
significant
mean
differ-
ence
between
CT
images
and
the
hoof-spe-
cific
radiographic
projections
for
number
was
found,
but
this
was
not
the
case
for
depth.
We
propose
two
explanations
for
the
low
agreement:
firstly,
the
superimposition
of
the
distal
sesamoid
bone
over
other
structures
on
the
DPr-PaDiO
projection
li-
mits
the
evaluation
of
these
small
anatomi-
cal
structures,
which,
however,
show
a
great
variation
in
their
morphology
(12);
and
secondly,
the
synovial
invaginations
are
situated
in
the
groove
between
the
distal
ar-
ticular
and
flexor
border
of
the
distal
sesa-
moid
bone,
making
it
very
difficult
to
de-
fine
their
most
distal
aspect
(19).
A
potential
limitation
of
this
study
could
be
the
absence
of
a
palmaroproxi-
mal-palmarodistal
oblique
projection,
which
permits
evaluation
of
the
distal
sesa-
moid
bone
without
superimposition.
However,
this
projection
only
allows
evalu-
ation
of
the
number
and
width
of
the
distal
border
synovial
invaginations,
and
was
therefore
not
included
in
our
study
(2).
We
conclude
that
our
hypothesis
can
be
rejected
and
the
radiographic
projection
with
hoof-specific
angle
may
not
be
con-
sidered
useful
in
the
evaluation
of
the
number
and
depth
of
distal
border
synovial
invaginations
of
the
distal
sesamoid
bone.
Computed
tomography
is
a
very
sensitive
technique
for
subtle
bone
changes
and
further
investigation
is
needed
to
evaluate
the
clinical
significance
of
the
changes
at
the
level
of
the
distal
sesamoid
bone
(es-
pecially
related
to
the
number
and
depth
of
the
synovial
invaginations).
Further
re-
search
is
required
to
investigate
the
clinical
importance
of
these
abnormal
synovial
in-
vaginations
since
their
presence
may
have
consequences
for
the
horse
and
its
owner.
Acknowledgements
The
authors
thank
Kim
Claus
and
Marnix
Verdonck
for
the
technical
assistance.
Conflict
of
interest
No
conflicts
of
interest
have
been
declared.
Source
of
funding
This
study
was
financed
by
the
"Bijzonder
Onderzoeksfonds",
Ghent
University.
References
1.
Colles
CM,
Hickman
J.
The
arterial
supply
of
the
navicular
bone
and
its
variations
in
navicular
dis-
ease.
Equine
Vet
J
1977;
9:
150-154.
2.
Rose
RJ,
Taylor
BJ,
Steel
JD.
Navicular
disease
in
the
horse:
an
analysis
of
seventy
cases
and
assessment
of
Vet
Comp
Orthop
Traumatol
6/2012
Schattauer
2012
S.
Claerhoudt
et
al.:
Morphology
of
synovial
invaginations
of
the
equine
distal
sesamoid
bone
459
a
special
radiographic
view.
J
Eq
Med
Surg
1978;
2:
492-497.
3.
Colles
CM.
Interpreting
radiographs
1:
The
foot.
Equine
Vet
J
1983;
15:
297-303.
4.
MacGregor
CM.
Radiographic
assessment
of
na-
vicular
bones,
based
on
changes
in
the
distal
nu-
trient
foramina.
Equine
Vet
J
1986;
18:
203-206.
5.
Pool
RR,
Meagher
DM,
Stover
SM.
Pathophysiology
of
navicular
syndrome.
Vet
Clinics
of
North
Am-
equine
Pract
1989;
5:
109-129.
6.
Kaser-Hotz
B,
Ueltschi
G.
Radiographic
appearance
of
the
navicular
bone
in
sound
horses.
Vet
Radiol
Ultrasound
1992;
33:
9-17.
7.
Wright
IM.
A
study
of
118
cases
of
navicular
dis-
ease:
radiological
features.
Equine
Vet
J
1993;
25:
493-500.
8.
Seyrek-Intas
D,
Tellhelm
B,
Reckels
FL
Interpre-
tation
and
diagnostic
value
of
some
radiological
findings
in
the
navicular
bone.
Pferdeheilkunde
1999;
15:
406-418.
9.
Robert
C,
Valette
JP,
Denoix
JM.
Correlation
be-
tween
routine
radiographic
findings
and
early
rac-
ing
career
in
French
Trotters.
Equine
Vet
J
Suppl
2006;
36:
473-478.
10.
Dik
KJ.
Diagnostische
beeldvorming
[Diagnostic
Imaging].
In:
Sloet
van
Oldruitenborgh-Ooster-
baan
MM,
Barneveld
A,
Van
Den
Belt
AL
editors.
De
Veterinaire
Keuring
van
het
Paard
[The
Purchase
Examination
of
the
Horse].
Leeuwarden:
Libre
BV;
2007.
pg.
85-112.
11.
Claerhoudt
S,
Bergman
HJ,
van
der
Veen
H,
et
al.
Differences
in
the
morphology
of
distal
border
syn-
ovial
invaginations
of
the
distal
sesamoid
bone
in
the
horse
as
evaluated
with
CT
compared
to
radi-
ography.
Eq
Vet
J
2012,
in
press.
doi:
10.1111/j.2042-3306.2012.00547.x.
12.
Claerhoudt
S,
Bergman
HJ,
van
der
Veen
H,
et
al.
Computed
tomographic
morphology
of
the
synov-
ial
invaginations
of
the
distal
sesamoid
bone
of
the
horse.
Anat
Histol
Embryol
2011;
40:
55-60.
13.
Murray
RC,
Blunden
TS,
Schramme
MC,
et
al.
How
does
magnetic
resonance
imaging
represent
histo-
logic
findings
in
the
equine
digit?
Vet
Radiol
Ultra-
sound
2006;
47:
17-31.
14.
Groth
AM,
May
SA,
Weaver
MP,
et
aL
Intra-
and
in-
terobserver
agreement
in
the
interpretation
of
na-
vicular
bones
on
radiographs
and
computed
to-
mography
scans.
Equine
Vet
J
2009;
41:
124-129.
15.
Widmer
WR,
Buckwalter
KA,
Fessler
JF,
et
aL
Use
of
radiography,
computed
tomography
and
magnetic
resonance
imaging
for
evaluation
of
navicular
syn-
drome
in
the
horse.
Vet
Radiol
Ultrasound
2000;
41:
108-116.
16.
Dyson
S,
Murray
R,
Blunden
T,
et
al.
Current
con-
cepts
of
navicular
disease.
Equine
Vet
Educ
2006;
18:45-56.
17.
Landis
JR,
Koch
GG.
The
measurement
of
observer
agreement
for
categorical
data.
Biometrics
1977;
33:
159-174.
18.
Altman
DG.
Some
common
problems
in
medical
research.
In:
Practical
statistics
for
medical
research.
London:
Chapman
&
Hall;
1991.
pg.
396-439.
19.
Rijkenhuizen
ABM,
Nemeth
F,
Dik
KJ,
et
al.
Devel-
opment
of
the
navicular
bone
in
foetal
and
young
horses,
including
the
arterial
supply.
Equine
Vet
J
1989;
21:
405-412.
C
Schattauer
2012
Vet
Comp
Orthop
Traumatol
6/2012