|
Presented at NACOB
98:
North
American Congress
on Biomechanics
Canadian Society for Biomechanics - American Society of
Biomechanics
University of Waterloo
Waterloo, Ontario, Canada
August 14-18, 1998
|
|
RELIABILITY AND VALIDITY OF FIRST
METATARSOPHALAGEAL JOINT
ORIENTATION, MEASURED WITH AN
ELECTROMAGNETIC TRACKING DEVICE
B.R. Umberger 1 , D.
Nawoczenski 2 , J.
Baumhauer 1
1 Department of
Orthopaedics,
University of Rochester Medical Center,
Rochester, NY 14642
2 Department of Physical
Therapy, Ithaca College-Rochester
Campus, Rochester, NY 14623
INTRODUCTION
Adequate motion of the first
metatarsophalageal
joint (FMTP) is an important component
of normal
ambulation. Researchers have relied on
planar
radiography, cinematography, and
videography, or
goniometry to measure FMTP motion.
Presently,
electromagnetic technology can be used
to monitor
3-D position and orientation of small
sensors,
attached to anatomical structures, with
respect to a
transmitter. Electromagnetic tracking
devices should
be ideal for measurement of FMTP
orientation,
however, to our knowledge the only
reported
application has been in an isolated
hallux-first
metatarsal preparation (Ahn et al.,
1997).
REVIEW AND THEORY
The sagittal plane motion of the FMTP
has been the
topic of a number of investigations
(e.g. Buell et al.,
1988; Hopson et al., 1995; Joseph,
1954). These
researchers have relied on the planar
techniques
mentioned above. Electromagnetic
tracking systems,
on the other hand, allow for data
collection in 3-D,
and have been reported to be useful for
kinematic
measurement (An et al., 1988). These
devices can
also be used for spatial digitization of
anatomical
landmarks, from which anatomically-based
reference
frames can be defined. By extracting the
sagittal
plane motion of the FMTP from 3-D
orientation
data, referred to an anatomical
reference frame, one
should avoid many of the errors inherent
in planar
motion analysis. The purpose of the
current
investigation was to establish the
reliability and
validity of measurement of the sagittal
plane
orientation of the FMTP using the Flock
of Birds
(FOB) electromagnetic tracking device
(Ascension
Technology). A cadaveric model was used
during
simulated clinical tests of FMTP motion.
PROCEDURES
Five fresh, frozen, cadaver foot
specimens (2 right
feet, 3 left feet) were used to
determine the reliability
and validity of FMTP measurements made
with the
FOB. Orientation was measured during
simulated
clinical tests of FMTP motion. The
clinical tests
were passive range of motion during
simulated
weight-bearing (PROM) and heel rise over
a fixed
hallux (HEEL). Specimens were prepared
in a
manner similar to that described by
Kitaoka et al.
(1995). The shank was transected at the
middle
third, and the most proximal portions of
the tibia and
fibula were cemented in a small pot with
polymethyl
methacrylate. Due to possible metallic
interference, a
testing jig was designed of wood and
nylon bolts.
The testing jig fixed the specimen in
place, and
allowed the application of an axial load
to simulate
weight-bearing, as well as, the setting
of mechanical
stops to control the motion of the
specimen.
The PROM trials were performed by
setting the
mechanical stop on the jig to just
inside the range of
maximal dorsiflexion of the specimen to
be tested.
The hallux was moved from the base of
the jig to the
mechanical stop while orientation data
were
collected. Three trials were performed
for each
testing condition, to establish
test-retest reliability.
The HEEL trials were performed by
removing the
axial load on the jig, and raising and
locking the top
platform of the jig. The specimen was
then rotated
around the hallux at the FMTP while the
hallux
remained on the base of the jig. The
specimen was
rotated around the hallux until the pot
contacted the
top platform of the jig. Marks on the
specimens and
jig were used to ensure consistent
beginning and
ending points of motion. Orientation of
the proximal
hallux relative to the first metatarsal
at the
mechanical stop was tracked by attaching
the
electromagnetic transmitter to the
testing jig and
sensors to the proximal hallux and first
metatarsal.
The receiver on the hallux was placed on
the
proximal, mediodorsal aspect of the
proximal
phalanx, medial to the extensor hallucis
longus
tendon. The receiver on the metatarsal
was placed on
the mediodorsal aspect of the diaphysis,
avoiding the
abductor hallucis and the tendon of the
extensor
hallucis longus.
After testing with the sensors applied
to the skin
with double sided adhesive, the sensors
were
anchored to the bone with nylon screws
through
small holes in the sensor casings. The
three tests
were repeated utilizing the same
mechanical stop
positions as for the skin-based trials.
Orientation
data were collected at 100 Hz and
smoothed using a
fourth-order Butterworth digital filter
with a 6 Hz
cutoff frequency. Dextral,
anatomically-based, local
coordinate systems were established for
the first
metatarsal and proximal hallux (Figure
1) by
digitizing anatomical landmarks on each
segment
from which unit vectors representing the
orthogonal
axes where constructed. A dextral global
coordinate
system was also established, with the
origin located
at the center of the transmitter.
Absolute sensor
orientation data were transformed using
matrices
describing the orientation of the hallux
local
coordinate system relative to an initial
coincident
alignment with the metatarsal local
coordinate
system (Craig, 1989). A Cardan angle
system of
three ordered rotations (X-Y-Z) was used
to extract
angular information of the proximal
hallux relative
to the first metatarsal.
Figure 1
Anatomical local coordinate systems
Although three-dimensional rotation
information
was calculated for FMTP motion, data
analysis was
restricted to the clinically relevant
sagittal plane
motion (X-axis rotation). Reliability
was determined
using intraclass correlation. Validity
was determined
using paired t-tests to uncover
measurement
differences. The skeletal-based
measurements were
assumed to represent actual motion of
the FMTP. An
alpha level of .05 was utilized.
RESULTS AND DISCUSSION
Means, standard deviations, and P values
of joint
orientations are reported in Table 1.
Reliability of all
three tests across both measurement
conditions was
excellent (intraclass r > .99). The
three orientation
values used to calculate each intraclass
r were
reduced to a mean value for further
analysis. The
mean differences in measurement of FMTP
orientation between the skin-based and
skeletal-
based techniques was -0.3 degrees for
the PROM trials, and
-0.4 degrees for the HEEL trials. None
of the differences
were statistically significant
| Variable |
Skin-based |
Skeletal-based |
P value |
| PROM |
44.2 ±
13.2 |
43.9 ± 11.4
|
.86 |
| HEEL |
46.3 ± 10.2
|
46.7 ± 9.7
|
.83 |
Table 1: FMTP joint orientation (in
degrees) with
skin-based and skeletal-based sensor
application
DISCUSSION
Although the FOB has been shown to be
accurate in
a non-biological setting (Milne et al.,
1996), the
reliability and validity of measurements
of FMTP
motion made with the FOB had not been
established.
The reliability of measurements made
with the FOB
was excellent. Each measurement was
repeated three
times, and reliability was assessed
using intraclass
correlation. Differences between
measurements made
with skin-based and skeletal-based
application of the
sensors were generally small, and were
comparable
to findings for similar measurements
made in the
wrist (Ishikawa et al., 1997). We feel
the FOB
electromagnetic tracking device may be
confidently
used for measurement of FMTP kinematics,
provided
that care is taken in selecting the
sites for sensor
application as described in the
procedures section.
REFERENCES
Ahn T. et al. Foot Ankle Int, 18, 170-4,
1997.
An K. et al. J Biomech, 21, 613-20,
1988.
Buell T. et al. J Am Podiatr Med Assoc,
78, 439-48,
1988.
Craig J. Introduction to Robotics,
15-53, Addison-
Wesley Company, 1989.
Hopson M. et al. J Am Pod Med Assoc, 85,
198-204,
1995.
Ishikawa J. et al. J Biomech, 30,
1183-6, 1997.
Joseph J. J Bone Joint Sur, 36, 450-7,
1954.
Kitaoka H. et al. Foot Ankle Int, 16,
492-9, 1995.
Milne A. et al. J Biomech, 29, 791-3,
1996.
