Effect of dorsopalmar projection obliquity on radiographic measurement of distal phalangeal rotation angle in horses with laminitis


Koblik, P.D.; O'Brien, T.R.; Coyne, C.P.

Journal of the American Veterinary Medical Association 192(3): 346-349

1988


The effect of dorsopalmar projection obliquity on calculation of distal phalangeal rotation (DPR) angle was determined in 8 feet obtained from 5 horses that had been euthanatized because of laminitis. A true lateromedial view of each foot served as a reference, with additional views taken at 5, 10, 15, and 20 degrees of x-ray tube head rotation in both a dorsal and in a palmar direction. Projection obliquity resulted in consistent underestimation of DPR angle. Where projection obliquity exceeded 10 degrees, there were significant (P less than 0.05) differences in DPR angle. The magnitude of underestimation appeared to be independent of severity of rotation. A radiographic variable, the mean normalized intercondylar distance, was defined and measured on 13 feet so that the angle of projection obliquity could be estimated on routine clinical studies.

Effect
of
dorsopalmar
projection
obliquity
on
radiographic
measurement
of
distal
phalangeal
rotation
angle
in
horses
with
laminitis
Philip
D.
Koblik,
DVM,
MS;
Timothy
R.
O'Brien,
DVM,
PhD;
Cody
P.
Coyne,
DVM,
PhD
Summary:
The
effect
of
dorsopalmar
projection
obliq-
uity
on
calculation
of
distal
phalangeal
rotation
(DPR)
angle
was
determined
in
8
feet
obtained
from
5
horses
that
had
been
euthanatized
because
of
laminitis.
A
true
lateromedial
view
of
each
foot
served
as
a
reference,
with
additional
views
taken
at
5,
10, 15,
and
20°
of
x-ray
tube
head
rotation
in
both
a
dorsal
and
in
a
palmar
direction.
Projection
obliquity
resulted
in
consis-
tent
underestimation
of
DPR
angle.
Where
projection
obliquity
exceeded
10°,
there
were
significant
(P<0.05)
differences
in
DPR
angle.
The
magnitude
of
underesti-
mation
appeared
to
be
independent
of
severity
of
rotation.
A
radiographic
variable,
the
mean
normalized
intercondylar
distance,
was
defined
and
measured
on
13
feet
so
that
the
angle
of
projection
obliquity
could
be
estimated
on
routine
clinical
studies.
p
almar
rotation
of
the
distal
phalanx
often
devel-
ops
as
a
part
of
equine
laminitis
syndrome.'
Radiography
traditionally
has
been
used
to
assess
the
presence
and
extent
of
the
rotation.
It
has
been
reported
that
the
degree
of
distal
phalangeal
rotation
(DPR)
can
be
used
to
predict
the
likelihood
for
return
to
function
after
resolution
of
laminitis.
2
Horses
with
less
than
5.5°
rotation
returned
to
previous
level
of
athletic
performance,
whereas
horses
with
greater
than
11.5°
rotation
either
had
persistent
lameness
sufficient
to
preclude
return
to
previous
level
of
performance
or
were
euthanatized
as
a
result
of
the
severity
of
the
laminitis.
Horses
with
intermediate
DPR
angles
could
still
perform
athletically,
but
at
a
lower
level
of
performance.
Despite
efforts
to
obtain
accurate
lateromedial
radiographs
of
the
foot,
some
obliquity
usually
is
present,
particularly
if
the
study
is
performed
on
a
patient
reluctant
to
bear
full
weight
on
the
affected
limbs.
The
objective
of
the
study
reported
here
was
to
assess
the
effect
of
From
the
Department
of
Surgery,
School
of
Veterinary
Medicine,
Tufts
University,
North
Grafton,
MA
01536
(Koblik),
and
the
Department
of
Radiological
Sciences,
School
of
Veterinary
Medicine,
University
of
California,
Davis,
CA
95616
(O'Brien,
Coyne).
Dr.
Koblik's
present
address
is
Department
of
Radiological
Sciences,
School
of
Veterinary
Medicine,
University
of
California,
Davis,
CA
95616.
dorsopalmar
radiographic
projection
obliquity
on
measured
DPR
angle
in
feet
obtained
from
horses
that
had
been
euthanatized
as
a
result
of
having
laminitis.
Materials
and
methods
Eight
feet
were
obtained
from
forelimbs
of
5
horses
necropsied
at
the
New
England
Veterinary
Medical
Center,
Tufts
University,
and
the
Veterinary
Medical
Teaching
Hospital,
University
of
California,
Davis.
Laminitis
had
been
diagnosed
clinically
and
radiographically
in
each
of
the
horses.
Each
foot
was
identified
as
left
or
right
and,
with
the
foot
in
a
normal
position
relative
to
the
ground,
was
radio-
graphed
with
the
primary
x-ray
beam
directed
lateral
to
medial
and
centered
at
the
level
of
the
coronary
band.
A
40-inch
focus-film
distance
was
maintained
on
all
views.
The
reference
lateromedial
projection
(ie,
of
obliquity)
was
obtained
by
positioning
the
x-ray
tube
head
relative
to
the
foot
in
such
a
manner
that
the
condyles
of
the
middle
phalanx
(P2)
were
superimposed.
Additional
views
were
then
taken
with
the
x-ray
tube
head
rotated
5,
10, 15,
and
20°
in
a
dorsal
and
in
a
palmar
direction.
The
angles
of
obliquity
were
obtained
by
aligning
the
beam-
centering
light
with
a
series
of
protractor
angles
transcribed
on
a
piece
of
paper
placed
below
the
foot
(Fig
1).
The
DPR
angle
was
calculated
by
a
method
described
by
Stick
et
al
(Fig
2).
2
An
additional
variable,
the
normalized
mean
intercon-
dylar
distance
(NMID),
was
calculated
in
an
effort
to
estimate
projection
obliquity
from
the
amount
of
parallax
(separation
of
the
medial
and
lateral
con-
dyles
of
P2)
on
each
view.
Distance
between
the
dorsal
margins
and
between
the
palmar
margins
of
the
2
condyles
were
first
averaged
and
then
divided
by
the
mean
dorsopalmar
thickness
of
the
condyles
to
correct
for
differences
in
P2
size
and
magnification
(Fig
2).
Similar
NMID
measurements
were
deter-
mined
on
5
additional
fore
feet
obtained
as
necropsy
specimens
from
horses
with
no
history
or
clinical
evidence
of
laminitis.
A
repeated
measures
analysis
of
variance
(ANovA)a
was
used
to
assess
the
relationship
between
'BMDP
2V,
BMDP
statistical
software,
Los
Angeles,
Calif.
346
Reports
of
Original
Studies
JAVMA,
Vol
192,
No.
3,
February
1,
1988
Figure
1—Drawing
of
foot
positioned
fo
r
true
lateromedial
study.
Primary
.x-ra
yy
beam
is
centered
at
the
coronary
ban
Angle
of
x-ray
tube
head
rotation
is
determined
by
position
of
collimator
cross-hair
relative
to
protractor
beneath
the
hoof
as
Figure
2—Radiogr
h
(A)
and
line
draw-
ing
(B)
of
a
15°
dorsolateral-palmaro-
medial
oblique
view
of
a
laminitic
foot.
Distal
phalangeal
rotation
angle
is
deter-
mined
by
subtracting
angle
of
the
hoof
wall
relative
to
the
ground
(1)
from
the
angle
of
the
dorsal
cortex
of
the
distal
phalanx
relative
to
the
ground
(2).
Nor-
malized
mean
intercondylar
distance
is
determined
by
dividing
the
average
inter-
condylar
distances
(1C
-1-13
1
2)
by
the
mean
condylar
widths
(jA
+B)n)
B
JAVIAA,
Vol
192,
No.
February
1,
1988
Reports
of
Original
Studies
347
171-1)
141-1)
1010)
51
-
4)
181-2)
141-6)
81-1)
41-1)
151-3)
131-2)
71-3)
41-
5)
171-3)
141-
6)
71-2)
31
-
2)
0.4
0.2
0.2
0.4
OBLIQUITY
10
15
20
palmar
20
15 10
AVERA
G
E
NMID
dorsal
Table
1-Angles
of
distal
phalangeal
rotation
measured
at
various
degrees
of
projection
obliquity
Dorsal
Palmar*
Foot
20°
15°
10°
1
151-34
141-4)
171-1)
1810)
2
141-1)
1510)
131-2)
141-1)
3
51-5)
61-4)
91-1)
91-1)
4
51-4)
71-
2)
71
-
2)
81-1)
5
171-3)
171-
3)
191-1)
2010)
6
171-3)
171
-
3)
181-
2)
191-1)
7
61-3)
51-4)
51-4)
61-3)
8
31-2)
41-1)
510)
41-1)
(-3.01
(-1.61
(-
1.0)
LM
10°
15°
20°t
18
1810)
171-1)
15
1510)
15101
10
1010)
1010)
9
71-2)
71
-
2)
20
191-1)
181-2)
20
171-3)
161-4)
9
71-2)
81-1)
5
510) 510)
-
1.0)
1-1.2)
(-2.0)
(
-3.215
Direction
of
x-ray
tube
head
rotation.
tAmount
of
projection
obliquity
in
degrees.
*DPR
angle
in
degrees
(change
in
DPR
angle
in
degrees
relative
to
on
projection).
§Average
change
in
DPR
angle
for
each
projection;
bold
values
were
significantly
different
(p<.05)
than
the
LM
view.
Lm
=
true
lateromedial
view.
projection
obliquity
and
DPR
angle
and
the
relation-
ship
between
NMID
value
and
projection
obliquity.
This
analysis
allowed
for
a
possible
dependence
between
the
results
obtained
on
a
given
hoof
at
different
angles
of
projection
obliquity.
A
repeated
measures
ANOVA
also
was
performed
to
assess
whether
the
relationship
between
DPR
and
projection
obliquity
was
linear
and
symmetric.
The
linearity
of
the
relationship
between
NMID
and
projection
obliq-
uity
was
studied
by
use
of
an
orthogonal
polynomial
decomposition
after
the
repeated
measures
ANOVA.
Results
Distal
phalangeal
rotation
angles
for
the
various
radiographic
projections
are
listed
in
Table
1.
In
each
case,
the
largest
DPR
angle
was
associated
with
the
true
lateromedial
projection,
although
in
5
of
the
8
feet
the
same
angle
was
calculated
at
other
projections.
Angles
of
pedal
bone
rotation
deter-
mined
from
projection
angles
greater
than
in
the
dorsal
direction
and
projection
angles
greater
than
10°
in
the
palmar
direction
were
significantly
differ-
ent
(P<0.05)
from
the
angle
obtained
on
the
true
lateromedial
projection.
Repeat
application
of
ANOVA,
with
an
additional
constraint
that
the
rela-
tionship
between
DPR
and
projection
obliquity
is
Table
2-Normalized
mean
middle
phalangeal
intercondylar
distances
at
various
degrees
of
projection
obliquity
Dorsal
Palmer*
Foot
20°
15°
10°
ur
10°
15°
20°1
.
Laminitis
1
0.36
0.30
0.19
0.11
0
0.08 0.18
0.31
0.38
2
0.36
0.30
0.14
0.09
0
0.09
0.19
0.31
0.38
3
0.30
0.19
0.07
0.01
0
0.13
0.28
0.30
0.38
4
0.31
0.25
0.13
0.08
0
0.11
0.25
0.33
0.39
5
0.36
0.27
0.18
0.08
0
0.15
0.26
0.34
0.39
6
0.38
0.32
0.24
0.13
0
0.09
0.18
0.25
0.34
7
0.40
0.31
0.24
0.10
0
0.08
0.16
0.26 0.36
8
0.40
0.31
0.23
0.11
0
0.06
0.18
0.26
0.34
Clinically
normal
9
0.29
0.20
0.12
0.07
0
0.14 0.24
0.34
0.42
10
0.38
0.25
0.18
0.08
0
0.06
0.18
0.27
0.36
11
0.35
0.31
0.21
0.14
0
0.05
0.11
0.25
0.35
12
0.42
0.34
0.22
0.10
0
0.08 0.18
0.29
0.42
13
0.35
0.30
0.18
0.06
0
0.12
0.25
0.34
0.41
*Direction
of
x-ray
tube
head
rotation.
tAmount
of
projection
obliquity
in
degrees.
linear
and
symmetrical
with
respect
to
the
true
lateromedial
projection,
accounted
for
95%
of
the
total
variability
of
the
data.
Effects
of
dorsal
vs
palmar
projection
obliquity
on
DPR
angle
were
not
significantly
different
and
were
independent
of
the
magnitude
of
pedal
rotation
measured
on
true
lateromedial
views.
Normalized
mean
intercondylar
distances
for
the
8
laminitic
feet
and
5
normal
feet
are
listed
in
Table
2.
There
was
a
significant
linear
relationship
(P<0.001)
between
NMID
and
degree
of
projection
obliquity,
whereas
all
other
polynomial
terms
were
insignificant.
The
line
of
regression
between
projec-
tion
obliquity
(Y)
vs
average
NMID
(X)
is
described
by
the
formula
Y
=
53.3
X;
R
2
=
1.00
(Fig
3).
Ex-
amination
of
the
radiographs
was
not
a
reliable
method
to
determine
whether
the
tube
head
had
been
moved
in
a
dorsal
or
palmar
direction.
Discussion
Even
if
the
dorsal
hoof
wall
and
the
distal
phalanx
were
concentric
cylinders
or
cones,
projec-
tion
obliquity
still
would
affect
DPR
angle,
because
the
plane
of
tube
head
rotation
is
not
perpendicular
to
the
long
axis
of
the
hoof.
Hence,
differential
Figure
3-Scatterplot
showing
linear
relationship
between
mean
NMID
vs
individual
dorsal
and
palmar
projection
obliquity
angles.
Regression
equation
for
this
line
is
obliq-
uity
=
53.3
X
NMID.
348
Reports
of
Original
Studies
JAVMA,
Vol
192,
No.
3,
February
1,
1988
Figure
4—Computed
tomographic
images
of
a
normal
foot
(A)
and
a
Ituninitic
foot
(B)
with
palmar
rotation
of
the
distal
phalanx.
Notice
that
the
distal
phalanx
and
dorsal
hoof
wall
are
elliptical
in
s
and
not
concentric
in
both
fart.
Notice
also
that
there
is
lateral
rotation
(arrow)
of
the
distal
phalanx
in
the
laminitic
foot
exaggerating
its
eccentricity
with
aspect
to
the
hoof
wall.
MED
=
medial;
MT
=
lateral.
magnification
would
alter
the
measured
distance
between
the
hoof
wall
and
the
parietal
cortex
of
the
distal
phalanx
in
the
toe
region
relative
to
this
distance
in
the
area
adjacent
to
the
extensor
process.
Further
distortion
is
created
by
the
fact
that
the
hoof
wall
and
the
distal
phalanx
are
not
simple
geometric
shapes
and
are
not
concentric.
As
shown
by
a
cross-sectional
tomographic
image
of
a
normal
foot
(Fig
4),
the
distal
phalanx
has
an
irregular
elliptical
shape.
The
dorsal
hoof
wall
also
is
elliptical
in
shape
and
can
vary
with
trimming.
The
distal
phalanx
is
closer
to
the
dorsal
hoof
wall
in
the
quarter
region
than
it
is
in
the
toe
region.
Palmar
rotation
of
the
distal
phalanx
exaggerates
this
dis-
parity.
Lateral
rotation
of
the
distal
phalanx
relative
to
the
hoof
wall
also
may
accompany
palmar
rotation
(Fig
4).
This
combination
of
factors
creates
complex
planar
imaging
geometry
and
the
common-sense
conclusion
that
projection
obliquity
is
likely
to
affect
calculation
of
DPR
angle.
In
general,
the
greater
the
projection
obliquity
in
either
a
dorsal
or
palmar
direction,
the
greater
the
inaccuracy
of
DPR
angle
calculation.
The
amount
of
underestimation
of
DPR
angle
for
the
various
projections
was
independent
of
severity
of
palmar
pedal
rotation.
This
was
a
somewhat
unexpected
result
and
indicated
that
inaccuracies
attributable
to
projection
obliquity
will
be
more
problematic
for
feet
with
minor
rotation
(eg,
feet
3,
4,
7,
8,
Table
1).
Results
of
this
study
indicated
that
projection
obliquity
can
cause
significant
underestimation
of
DPR
angle
in
horses
with
laminitis,
particularly
those
in
which
the
degree
of
rotation
is
mild.
Use
of
cm
angle
to
predict
probability
for
return
to
function
requires
careful
attention
to
the
quality
of
the
radiographic
image
from
which
the
angle
was
determined.
Although
somewhat
complex,
we
chose
NMID
as
the
most
suitable
variable
to
predict
projec-
tion
obliquity
because
it
should
be
unaffected
by
size
of
the
hoof
or
by
radiographic
magnification.
The
extremely
close
relationship
demonstrated
be-
tween
NMID
and
obliquity
angle
supports
this
con-
clusion.
The
linear
regression
equation
describing
this
relationship
was
reported
citing
NM1D
as
the
independent
variable,
as
would
be
the
case
for
clinical
studies.
A
reasonable
criterion
for
acceptance
of
clinical
radiographic
studies
would
be
a
NMID
value
less
than
0.15.
This
would
ensure
that
projec-
tion
obliquity
would
be
no
greater
than
8
to
10°.
The
calculated
DPR
angle,
for
this
degree
of
projection
obliquity,
would
not
be
expected
to
differ
signifi-
candy
from
that
obtained
on
a
perfect
lateromedial
radiograph.
References
1.
Coffman
JR.
Johnson
J1-1,
Finocchio
FJ,
et
al.
Biomechan-
ics
of
pedal
bone
rotation
in
equine
laminitis.
J
Am
Vet
Med
Assoc
1970;146:219-221.
2.
Stick
JA,
Jann
HW,
Scott
EA,
et
al.
Pedal
bone
rotation
as
a
prognostic
sign
in
taminitis
in
horses.
J
Am
Vet
Med
Assoc
1982;180:251-253.
JA
,
Vol
192,
No.
3,
February
1,
1988
Reports
of
Original
Studies
349