From the *Department of Orthopedic Surgery, Osaka University
Medicine, Suita, Osaka, Japan.

Received for publication October 17, 2016; accepted in revised form

This work was supported by Japan Society for the Promotion of Sc
Number JP15K10442.

Corresponding author: Kunihiro Oka, MD, PhD, Department of
Osaka University Graduate School of Medicine, 2-2, Yamada-oka, Su
Japan; e-mail: [email protected]

0363-5023/18/4302-0012$36.00/0
http://dx.doi.org/10.1016/j.jhsa.2017.07.028

182.e1 r � 2018 ASSH r Published by Elsevier, I

SCIENTIFIC ARTICLE

In Vivo 3-Dimensional Kinematics of Thumb

Carpometacarpal Joint During Thumb Opposition

Yohei Kawanishi, MD, PhD,* Kunihiro Oka, MD, PhD,* Hiroyuki Tanaka, MD, PhD,*
Kiyoshi Okada, MD, PhD,* Kazuomi Sugamoto, MD, PhD,* Tsuyoshi Murase, MD, PhD*

Purpose This study primarily aimed to demonstrate the screw-home rotation of the thumb car-
pometacarpal (CMC) joint and the function of surrounding ligaments during thumb oppositional
motion.

Methods A 3-dimensional kinematic analysis of the thumb CMC joint was conducted using data
derived from computed tomography of 9 healthy volunteers. Scans were obtained in the neutral
forearmandwrist position and the thumb inmaximumradial abduction,maximumpalmar abduction,
andmaximumopposition. Themovements of thefirstmetacarpal and the palmar and dorsal bases on
the trapezium during thumb oppositional motion from radial abduction through palmar abduction
were quantified using a coordinate system originating on the trapezium. In addition to the kinematic
analyses, the length of virtual ligaments, including the anterior oblique, ulnar collateral, dorsal radial,
dorsal central (DCL), and posterior oblique ligament (POL), were calculated at each thumb position.

Results From radial abduction to opposition of the thumb through palmar abduction, the first
metacarpal was abducted, internally rotated, and flexed on the trapezium. The palmar base of
the first metacarpal moved in the palmar-ulnar direction, and the dorsal base moved in the
palmar-distal direction along the concave surface of the trapezium. Although the DCL and
POL lengthened, the lengths of other ligaments did not change significantly.

Conclusions During thumb oppositionalmotion, internal rotation of the first metacarpal occurred,
with the palmar base rotating primarilywith respect to the dorsal base. TheDCLandPOLmaybe
strained in thumb functional positions.

Clinical relevance Kinematic variables indicated a screw-home rotation of the thumb CMC joint
and the contribution of the dorsal ligaments to the stability of the rotation on the pivot point.
(J Hand Surg Am. 2018;43(2):182.e1-e7. Copyright � 2018 by the American Society for
Surgery of the Hand. All rights reserved.)
Key words Dorsal ligaments, kinematics, osteoarthritis, 3-dimensional, thumb carpometacarpal
joint.

Graduate School of

July 26, 2017.

ience KAKENHI Grant

Orthopedic Surgery,
ita, Osaka 565-0871,

nc. All rights reserved

T HE THUMB CARPOMETACARPAL (CMC) joint is a
saddle-shaped and semiconstrained joint with
a wide range of motion. Thumb mobility thus

allows the performance of unique functions such as
pinching and grasping in tandem with other fingers.
During these movements, the thumb CMC joint is
proposed to be stabilized by a screw-home rotation
that allows the palmar beak of the first metacarpal
to lock into the palmar recess of the trapezium.1 The
anterior oblique ligament (AOL), 1 of the palmar

.

3-DIMENSIONAL KINEMATICS OF THUMB CMC JOINT 182.e2

ligaments of the thumb CMC joint, has been regarded
as the primary stabilizer of the joint,2 but recent
studies indicate that the AOL plays a secondary sta-
bilization role compared with the dorsal ligament
complex. An anatomical study showed a lack of deep
fibers of the AOL in 27 of 30 normal cadaveric hands
(90%), and tissues of the dorsal ligaments appeared
more organized, with greater cellularity and more
sensory innervation than the AOL.3 Several biome-
chanical studies have found that the dorsal ligaments
are stronger than the AOL.4e6 Taut dorsal ligaments
are proposed to function as stabilizers during screw-
home rotation and enable the palmar beak of the
first metacarpal to cantilever into the palmar recess of
the trapezium.1

The dorsal radial ligament (DRL) has been consid-
ered to be the strongest of the dorsal ligaments4 and
more important as a stabilizer6,7 of the thumb CMC
joint than the posterior oblique ligament (POL). How-
ever, the dorsal central ligament (DCL), which is the
thickest and shortest component of the dorsal ligament
complex,3 may also be a stabilizer. The contributions of
the DRL and DCL are unclear because previous studies
have not assessed these 2 ligaments separately.4e9

Evaluation of joint motion in 2 axes is difficult
with conventional 2-dimensional techniques,10e13 but
a markerless 3-dimensional analytical technique is
available for in vivo kinematic analysis of the thumb
CMC joint.14e17 Crisco et al14 described the orien-
tation of the functional axes of the thumb CMC joint
as a potentially stabilizing screw-home rotation.
Computed recruitment patterns of the AOL and DRL
indicated the importance of the DRL.8

We evaluated 3-dimensional thumb CMC joint
kinematics and virtual ligament length surrounding
the joint during thumb opposition, focusing on the
motion of the first metacarpal base because articular
interactions reflect the details of screw-home rotation
and the mechanical effects on the surrounding liga-
ments. The study objective was to use in vivo kine-
matics data to describe the screw-home motion of the
thumb CMC joint during thumb oppositional motion
and the function of each of the 3 dorsal ligaments.

METHODS
Nine male Japanese volunteers (mean age, 34.3 years;
range 32e42 years) with no history of trauma or
disease of the right upper extremity were included in
the analysis. The study protocol was approved by the
local institutional review board. The kinematics of the
thumb CMC joint during thumb motion included first
metacarpal movement on the trapezium and focused

J Hand Surg Am. r Vo

on the palmar and dorsal bases of the first metacarpal
bone. Movements were assessed 3-dimensionally and
quantified using a coordinate system originating on
the trapezium.

The right wrist was immobilized using a short-arm
thumb spica cast with the thumb in a position of
maximal palmar abduction in a plane perpendicular to
the palm and the wrist aligned in a neutral position. A
roll of 3-inch padding and 3-inch fiberglass cast ma-
terial (3M Company, Oakdale, MN) was used for each
wrist. Hands were scanned by low-radiation computed
tomography (CT) (scan time, 0.5 s; slice thickness,
1.25mm; 10mA; 120 kV) using a LightSpeedUltra 16
CT system (General Electric, Waukesha, WI) pro-
ducing one-thirtieth of the normal radiation dose.18

The subjects were in the prone position on the CT ta-
ble with their arms elevated over the head, the elbows
flexed at 90�, and the forearm in a neutral position
during the scan. After a CT scan with the maximum
palmar thumb abduction, the thumb portion of the
thumb spica cast was removed to allow the thumb to
abduct radially and oppose to the proximal palmar
crease of the little finger, which was defined as
maximum opposition. Each subject underwent a CT
scan with the thumb in 2 additional positions:
maximum radial abduction in the plane of the palm and
maximum opposition. Data were saved in the Digital
Imaging and Communications in Medicine format
and stored in a computer (Dell Precision M4600, 2.50
GHz/4G; Dell, Inc., Round Rock, TX). Contours of
the radius, ulna, carpal bones, and first metacarpal
3-dimensional surface generation of the bone cortex19

were created using commercial software (BV; Orthree
Co., Ltd., Osaka, Japan). The digital models were
visualized using software (BS; Orthree Co., Ltd.) that
enabled digital 3-dimensional measurements on the
computer.

Thumb CMC joint kinematics

Kinematic variables of the thumb CMC joint were
calculated by registering the bone in various positions,
which were then compared relative to an orthogonal
coordinate system originating on the trapezium
(Fig. 1A) as previously reported.20e22 The z axis,
whichwas nearly parallel to the central ridge, indicated
the radial (þ)/ulnar (�) direction and passed through
the top of the radial facet and the top of the ulnar facet.
The x axis, indicating the palmar (þ)/dorsal (�)
direction, ran perpendicular to the z axis and passed
through the midpoint of the dorsal surface. The y axis,
indicating the proximal (þ)/distal (�) direction, was
perpendicular to the x and z axes. Rotation around
the x axis indicated ulnar (þ)/radial (�) deviation,

l. 43, February 2018

FIGURE 1: Orthogonal reference system including the trapezium.AAblue broken line is the central ridge of the trapezial articular surface.
B The palmar and dorsal cusps (green circles) of the proximal articular surface of the first metacarpal in lateral and proximal views.

182.e3 3-DIMENSIONAL KINEMATICS OF THUMB CMC JOINT

rotation around the y axis indicated internal rotation
(þ)/external rotation (�), and rotation around the
z axis indicated flexion (þ)/extension (�). The posi-
tional changes in the first metacarpal relative to the
trapezium were calculated using Euler angles, which
quantify spatial joint motion using 3 sequential rota-
tions for interpreting motion in coordinate systems.17

The palmar and dorsal cusps of the proximal base of
the first metacarpal were assigned point values
(Fig. 1B), and the positions of these points in the
trapezium coordinate system were also calculated at
each position during thumb motion. The thumb CMC
joint motions were analyzed in the following order:
maximum radial position, palmar, and oppositional
position using the static CT data in these 3 positions.

Ligament length

Two palmar ligaments, the AOL and the ulnar collat-
eral ligament (UCL), and 3 dorsal ligaments, the
DRL, DCL, and POL, were studied. The origins and
insertions of these ligamentsweremanuallymarked on
the bone models as described by Ladd et al3 (Fig. 2).
The AOL attaches proximally on the palmar crest of
the trapezium and distally on the palmar beak of the
first metacarpal. The UCL attaches proximally on
the ulnar-palmar ridge of the trapezium and distally
on the palmar-ulnar edge of the first metacarpal base.
The DRL, DCL, and POL attach proximally on the
radial, center, and ulnar aspects of the dorsal tubercle of

J Hand Surg Am. r Vo

the trapezium and distally on the radial side, center,
and ulnar side of the dorsal aspect of the first meta-
carpal base, respectively. Avoiding bone penetration
in 3-dimensional space, all ligament paths were
modeled, and the length at each positionwas computed
as the shortest distance between the origin and the
insertion.8,23 Because this method could not account
for the effects of ligament tension, significant changes
in static length were defined as taut or loosened
ligaments.

Data analysis

Differences in measurements at opposition or palmar
abduction and at radial abduction were compared, and
ligament lengthening or shortening was expressed as
percentages. Data were expressed as means and stan-
dard error; 1-way analysis of variance was performed
to determine significant differences. A P value of less
than .05 was considered statistically significant.

RESULTS
Thumb CMC joint kinematics

During thumb movement from radial abduction to
opposition through palmar abduction, the first
metacarpal was abducted, internally rotated, andflexed
significantly on the trapezium (Fig. 3). During
movement from radial to palmar abduction, the first
metacarpal abducted 25.8� � 6.6� (P< .05), internally

l. 43, February 2018

FIGURE 2: A Palmar ligament models show the AOL (red) and UCL (orange). B Dorsal ligament models show the DRL (blue), DCL
(turquoise), and POL (light turquoise). C The estimated length values during thumb motion.

FIGURE 3: Motion of the first metacarpal during thumb opposition. The first metacarpal A abducted, B internally rotated, and C flexed
on the trapezium. Curved red arrows show the motion from radial abduction (blue bone) to palmar abduction (yellow bone), and the
curved blue arrows show the motion from palmar abduction to opposition (pink bone).

3-DIMENSIONAL KINEMATICS OF THUMB CMC JOINT 182.e4

rotated 16.8� � 4.0� (P< .05), and flexed 20.6� � 4.5�

(P < .05). During movement from radial abduction to
opposition, the first metacarpal abducted 26.6� � 6.6�

(P < .05), internally rotated 22.3� � 4.0� (P < .05),
and flexed 25.0� � 4.5� (P < .05).

During the sequence of thumb oppositional mo-
tion, the first metacarpal flexes and pronates with the
dorsal base as the center. The palmar side moved
significantly in the palmar-ulnar direction and the
dorsal side moved significantly in the palmar-distal
direction during movement from radial abduction to
opposition (Fig. 4A and B). From radial abduction to
opposition, the palmar cusp of the proximal articular
surface of the first metacarpal translated 2.5 � 0.7
mm in the palmar direction (P < .05), 1.3 � 0.6 mm
in the proximal direction (P ¼ .09), and 5.7 � 1.0
mm in the ulnar direction (P < .05). The dorsal

J Hand Surg Am. r Vo

cusp translated 3.1 � 0.8 mm in the palmar direction
(P < .05), 3.9 � 0.8 mm in the distal direction
(P < .05), and 0.5 � 0.7 mm in the ulnar direction
(P ¼ .7). From radial to palmar abduction, the palmar
cusp translated 2.3 � 0.7 mm in the palmar direction
(P < .05), 0.8 � 0.6 mm in the proximal direction
(P ¼ .4), and 5.0 � 1.0 mm in the ulnar direction
(P < .05). The dorsal cusp translated 2.4� 0.8 mm in
the palmar direction (P < .05), 3.5 � 0.8 mm in the
distal direction (P < .05), and 1.2 � 0.7 mm in the
ulnar direction (P ¼ .2).

Ligament length

Compared with the ligament length in radial abduc-
tion, the palmar ligaments shortened, while the dorsal
ligaments lengthened, in palmar abduction and op-
position (Fig. 2C). Although the length changes in the

l. 43, February 2018

FIGURE 4: Movement locus of the palmar and dorsal cusps of the proximal articular surface of the first metacarpal on the articular
surface of trapezium (green model) in A distal and B lateral views; blue, yellow, and red points indicate the cusps in radial abduction,
palmar abduction, and opposition, respectively. The dorsal cusp moves in the distal-palmar direction and the palmar cusp moves in the
palmar-ulnar direction in the palmar recess of thumb CMC joint during thumb oppositional motion from radial abduction. C These
results mean that, during thumb oppositional motion, the first metacarpal flexes and rotates internally with the dorsal base as the center.

182.e5 3-DIMENSIONAL KINEMATICS OF THUMB CMC JOINT

DCL and POL were significant, the palmar ligaments
and the DRL did not show any significant change
in length. The DCL lengthened 2.3� 0.5 mm (41.6%;
P < .05) and 2.1 � 0.5 mm (38.8%; P < .05), and the
POL lengthened 4.0 � 0.8 mm (47.9%; P < .05) and
3.8 � 0.8 mm (46.6%; P < .05) from radial to palmar
abduction and to opposition, respectively. The AOL
shortened 1.2� 0.6 mm (20.3%; P¼ .2) and 0.9� 0.6
mm (15.6%; P ¼ .4), the UCL shortened 0.6 � 0.5
mm (10%; P¼ .5) and 0.9� 0.5 mm (14.4%; P¼ .2),
and the DRL lengthened 0.4� 0.5 mm (5.6%; P¼ .7)
and 0.8 � 0.5 mm (12.8%; P ¼ .3) from radial to
palmar abduction and to opposition, respectively.

DISCUSSION
Screw-home rotation in the thumb CMC joint is pro-
posed to stabilize the thumb during movement. Taut
dorsal ligaments of the thumb CMC joint enable
rotation by a cantilever mechanism during thumb
functional movement.1 Although previous biome-
chanical studies have confirmed the importance of the
dorsal ligaments as stabilizers of the thumb CMC,4e7

the 3 dorsal ligaments have not been evaluated sepa-
rately. We analyzed in vivo 3-dimensional kinematics
of the thumb CMC joint and changes in length of 2
virtual palmar and 3 dorsal ligaments during thumb
opposition, while paying close attention to the motion
of the first metacarpal base. The aim was to obtain
kinematic evidence of screw-home rotation and the
involvement of the 3 dorsal ligaments.

The study limitations include the sample size and
character of the objectives. Although a larger number

J Hand Surg Am. r Vo

of participants may have resulted in different findings,
our analysis had adequate power of greater than 0.9 in a
post hoc power analysis (a ¼ 0.05). As all subjects in
this studywere Japanesemen aged 30 to 49 years, there
was no diversity of race, sex, and age. Consequently,
racial-, sex-, and age-related differences in thumb
CMC joint kinematics could not be clarified in this
study. Third, the angle of thefirst metacarpal relative to
the trapezium in each thumb position was not neces-
sarily absolutely the same in all participants as
shown in the standard error of the measurements. This
variability might have been affected by individual
differences in in vivo joint congruence24 and joint
laxity.21 Fourth, the study did not consider the influ-
ence of different fibers of each ligament because they
were calculated as a line between their anatomical
origin and did not have a width. Finally, we assessed
the thumb CMC joint only at maximum radial abduc-
tion, maximum palmar abduction, and maximum
thumb opposition. However, the thumb can move
intricately over a range of circumduction during these
functions. Therefore, there exists a possibility that the
thumb CMC joint has kinematics that were not
apparent in this analysis.

Although the first metacarpal internally rotated on
the trapezium during opposition from radial abduction
(Fig. 3B), the motion of the first metacarpal indicated
that the palmar base rotated primarily with respect to
the dorsal base in the palmar recess of thumb CMC
joint (Fig. 4C). Edmunds1 hypothesized a screw-home
rotation of the thumb CMC joint in which the palmar
base of the first metacarpal rotates in the palmar recess
of the joint. Recent kinematic studies found that the

l. 43, February 2018

3-DIMENSIONAL KINEMATICS OF THUMB CMC JOINT 182.e6

rotation is coupled with thumb flexion-extension and
abduction-adduction, and that internal rotation of the
first metacarpal is coupled with thumb flexion or
abduction.14,15 Kinematic variables in our study
visualized the screw-home rotation and supported the
results of previous studies because the first metacarpal
in our series also abducted and flexed on the trapezium
along with internal rotation (Fig. 3A, C).

Stable screw-home motion of the thumb CMC joint
can be facilitated by taut dorsal ligaments working as
a joint stabilizer during thumb oppositional motion.6

In this study, during thumb opposition from radial
abduction, the dorsal ligaments, except for the DRL,
lengthened, but the lengths of the 2 palmar ligaments
did not change significantly. If these dorsal ligaments
are presumed to remain taut as they lengthen, they can
be seen as contributing to a stable screw-home rota-
tion. Halilaj et al8 reported high in vivo recruitment of
the DRL, separated into radial, central, and ulnar
fibers that could be regarded as a taut ligament, with
radial fibers stabilizing adduction and the ulnar fiber
stabilizing both adduction and flexion. These radial
and ulnar fibers can be considered as the DRL and
DCL in our study, respectively, because the path of
each fiber in their study was nearly the same as that of
the ligaments in our study. Their ulnar fiber was the
shortest dorsal ligament, which is characteristic of the
DCL. Therefore, the DRL might not significantly
lengthen for abduction, and the DCL might signifi-
cantly lengthen for flexion of the thumb CMC joint.
The POL lengthened significantly during thumb
opposition from radial abduction, as previously
reported.9 The contribution of the POL as a stabilizer
of the thumb CMC joint may be smaller than that of
the DRL and DCL.6 However, the POL may
contribute to a stable screw-home rotation in combi-
nation with a taut DCL that increases in length during
thumb movement. To demonstrate additional details
of the changes in the ligaments surrounding the thumb
CMC joint during thumb motion, cadaver studies to
measure the tension of each ligament are necessary.
Because the location of these ligaments, especially
palmar ones, is variable, the information that can be
gained from the estimated length changes of virtual
ligaments in this and previous studies are limited.

The thumb CMC joint is commonly affected by
osteoarthritis, with the palmar portion being most
often involved.24e27 Although thumb CMC joint
osteoarthritis may be associated with AOL degener-
ation,27 palmar cartilage wear in the thumb CMC
joint can develop without AOL degeneration.28 Our
study revealed rotational motion of the first meta-
carpal on the trapezium with the dorsal side as a

J Hand Surg Am. r Vo

pivot. The cartilage on the palmar-ulnar portion of the
thumb CMC joint is the active contact area during
thumb function and is thinner than the cartilage in
other areas of the joint.29 The predominant rotation of
the palmar side of the first metacarpal base may be an
additional, direct cause of damage of the articular
cartilage and contribute to the development of thumb
CMC joint osteoarthritis. However, additional kine-
matic study comparing arthritic and normal hands is
needed to confirm our hypothesis of the pathogenesis
of thumb CMC joint osteoarthritis.

This study analyzed the in vivo 3-dimensional
kinematics of the thumb CMC joint and the length
changes in the surrounding ligaments during thumb
opposition. The rotational motion of the first meta-
carpal base on the trapezium predominantly occurred
on the palmar side with the dorsal side as a pivot,
indicating screw-home rotation with the possible
effect of the DCL and POL on stability during thumb
oppositional motion.

REFERENCES

1. Edmunds JO. Current concepts of the anatomy of the thumb
trapeziometacarpal joint. J Hand Surg Am. 2011;36(1):170e182.

2. Pellegrini VD Jr. Osteoarthritis of the trapeziometacarpal joint: the
pathophysiology of articular cartilage degeneration. I. Anatomy and
pathology of the aging joint. J Hand Surg Am. 1991;16(6):967e974.

3. Ladd AL, Lee J, Hagert E. Macroscopic and microscopic analysis of
the thumb carpometacarpal ligaments: a cadaveric study of ligament
anatomy and histology. J Bone Joint Surg Am. 2012;94(16):
1468e1477.

4. Bettinger PC, Smutz WP, Linscheid RL, Cooney WP III, An KN.
Material properties of the trapezial and trapeziometacarpal ligaments.
J Hand Surg Am. 2000;25(6):1085e1095.

5. Colman M, Mass DP, Draganich LF. Effects of the deep anterior
oblique and dorsoradial ligaments on trapeziometacarpal joint
stability. J Hand Surg Am. 2007;32(3):310e317.

6. Strauch RJ, Behrman MJ, Rosenwasser MP. Acute dislocation of the
carpometacarpal joint of the thumb: an anatomic and cadaver study.
J Hand Surg Am. 1994;19(1):93e98.

7. Van Brenk B, Richards RR, Mackay MB, Boynton EL.
A biomechanical assessment of ligaments preventing dorsoradial
subluxation of the trapeziometacarpal joint. J Hand Surg Am.
1998;23(4):607e611.

8. Halilaj E, Rainbow MJ, Moore DC, et al. In vivo recruitment patterns
in the anterior oblique and dorsoradial ligaments of the first carpo-
metacarpal joint. J Biomech. 2015;48(10):1893e1898.

9. Tan J, Xu J, Xie RG, Deng AD, Tang JB. In vivo length and changes
of ligaments stabilizing the thumb carpometacarpal joint. J Hand
Surg Am. 2011;36(3):420e427.

10. Cerveri P, De Momi E, Marchente M, et al. Method for the estimation
of a double hinge kinematic model for the trapeziometacarpal joint
using MR imaging. Comput Methods Biomech Biomed Engin.
2010;13(3):387e396.

11. Cerveri P, De Momi E, Marchente M, et al. In vivo validation of a
realistic kinematic model for the trapezio-metacarpal joint using an
optoelectronic system. Ann Biomed Eng. 2008;36(7):1268e1280.

12. Hollister A, Buford WL, Myers LM, Giurintano DJ, Novick A. The
axes of rotation of the thumb carpometacarpal joint. J Orthop Res.
1992;10(3):454e460.

l. 43, February 2018

182.e7 3-DIMENSIONAL KINEMATICS OF THUMB CMC JOINT

13. Miura T, Ohe T, Masuko T. Comparative in vivo kinematic analysis
of normal and osteoarthritic trapeziometacarpal joints. J Hand Surg
Am. 2004;29(2):252e257.

14. Crisco JJ, Halilaj E, Moore DC, Patel T, Weiss AP, Ladd AL. In vivo
kinematics of the trapeziometacarpal joint during thumb extension-
flexion and abduction-adduction. J Hand Surg Am. 2015;40(2):
289e296.

15. Crisco JJ, Patel T, Halilaj E, Moore DC. The envelope of physio-
logical motion of the first carpometacarpal joint. J Biomech Eng.
2015;137(10):101002.

16. Halilaj E, Rainbow MJ, Got C, et al. In vivo kinematics of the thumb
carpometacarpal joint during three isometric functional tasks. Clin
Orthop Relat Res. 2014;472(4):1114e1122.

17. Goto A, Leng S, Sugamoto K, Cooney WP III, Kakar S, Zhao K.
In vivo pilot study evaluating the thumb carpometacarpal joint during
circumduction. Clin Orthop Relat Res. 2014;472(4):1106e1113.

18. Oka K, Murase T, Moritomo H, Goto A, Sugamoto K, Yoshikawa H.
Accuracy analysis of three-dimensional bone surface models of the
forearm constructed from multidetector computed tomography data.
Int J Med Robot. 2009;5(4):452e457.

19. Lorensen W, Cline H. Marching cubes: a high-resolution 3D surface
construction algorithm. Comput Graph. 1987;21(4):163e169.

20. Cooney WP III, Lucca MJ, Chao EY, Linscheid RL. The kinesiology
of the thumb trapeziometacarpal joint. J Bone Joint Surg Am.
1981;63(9):1371e1381.

21. Halilaj E, Rainbow MJ, Got CJ, Moore DC, Crisco JJ. A thumb
carpometacarpal joint coordinate system based on articular surface
geometry. J Biomech. 2013;46(5):1031e1034.

J Hand Surg Am. r Vo

22. Wu G, van der Helm FC, Veeger HE, et al. ISB recommendation on
definitions of joint coordinate systems of various joints for the
reporting of human joint motion—part II: shoulder, elbow, wrist and
hand. J Biomech. 2005;38(5):981e992.

23. Marai GE, Laidlaw DH, Demiralp C, Andrews S, Grimm CM,
Crisco JJ. Estimating joint contact areas and ligament lengths from
bone kinematics and surfaces. IEEE Trans Biomed Eng. 2004;51(5):
790e799.

24. Pellegrini VD Jr. The ABJS 2005 Nicolas Andry Award: osteoar-
thritis and injury at the base of the human thumb: survival of the
fittest? Clin Orthop Relat Res. 2005;438:266e276.

25. Pellegrini VD Jr, Smith RL, Ku CW. Pathobiology of articular
cartilage in trapeziometacarpal osteoarthritis. I. Regional biochemical
analysis. J Hand Surg Am. 1994;19(1):70e78.

26. Pellegrini VD Jr, Smith RL, Ku CW. Pathobiology of articular
cartilage in trapeziometacarpal osteoarthritis. II. Surface ultrastruc-
ture by scanning electron microscopy. J Hand Surg Am. 1994;19(1):
79e85.

27. Pelligrini VD Jr. Osteoarthritis of the trapeziometacarpal joint: the
pathophysiology of articular cartilage degeneration. II. Articular wear
patterns in the osteoarthritic joint. J Hand Surg Am. 1991;16(6):
975e982.

28. Pellegrini VD Jr, Olcott CW, Hollenberg G. Contact patterns in the
trapeziometacarpal joint: the role of the palmar beak ligament.
J Hand Surg Am. 1993;18(2):238e244.

29. Ateshian GA, Ark JW, Rosenwasser MP, Pawluk RJ, Soslowsky LJ,
Mow VC. Contact areas in the thumb carpometacarpal joint. J Orthop
Res. 1995;13(3):450e458.

l. 43, February 2018