JHS(E) - TU Delft

466764
2012
JHS0010.1177/1753193412466764Journal of Hand SurgeryRaedt et al.
JHS(E)
Full length article
A three-dimensional analysis of
osteoarthritic changes in the thumb
carpometacarpal joint
The Journal of Hand Surgery
(European Volume)
0E(0) 1­–9
© The Author(s) 2012
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1753193412466764
jhs.sagepub.com
S. de Raedt
Quantitative Imaging Group, Delft University of Technology, The Netherlands, Orthopaedic Research Unit, Regional
Hospital Holstebro, Denmark and Orthopedic Research Department, Aarhus University Hospital, Denmark
M. Stilling
Orthopaedic Research Unit, Regional Hospital Holstebro, Denmark
M. van de Giessen
Quantitative Imaging Group, Delft University of Technology, Delft and Department of Biomedical Engineering and Physics,
Academic Medical Center, Amsterdam, The Netherlands
G. J. Streekstra
Department of Biomedical Engineering and Physics, Academic Medical Center, Amsterdam, The Netherlands
F. M. Vos
Quantitative Imaging Group, Delft University of Technology, Delft and Department of Radiology, Academic Medical Center,
Amsterdam, The Netherlands
T. B. Hansen
Orthopaedic Research Unit, Regional Hospital Holstebro, Holstebro, Denmark
Abstract
The purpose of this study is to gain a better understanding of the changes due to osteoarthritis (OA) occurring
in the thumb carpometacarpal (CMC) joint by comparing quantitative geometrical measurements in computed
tomography scans of healthy and pathological joints in various stages of OA. The measurements were (1) the
subluxation of the metacarpal on the trapezium, (2) distance from the scaphoid centre to the metacarpal base,
and (3) distance from the metacarpal base to the articulating surface of the trapezium. The three-dimensional
position of three characteristic points on the metacarpal, trapezium, and scaphoid were detected in each of
the 90 wrists we scanned. The distances between the points were compared by statistical analysis. With high
accuracy, we have been able to confirm and quantify that subluxation occurs in the dorso-radial direction.
A significant difference in trapezium height and joint space width was found between the OA and control
groups. The results indicate how to restore the centre of rotation in surgical treatment of OA with total joint
arthroplasty, but the clinical relevance of these findings has to be tested in further clinical studies.
Keywords
Osteoarthritis, trapeziometacarpal joint, trapezium, carpal bones, CT
Date received: 13th July 2011; revised: 10th October 2012; accepted: 14th October 2012
Introduction
The prehensile biomechanical function of the hand is
largely due to the complex saddle-shaped thumb carpometacarpal (CMC) or trapeziometacarpal (TM) joint,
Corresponding author:
Sepp de Raedt, Department of Orthopedics, Aarhus University
Hospital, Tage-Hansens Gade 2, Building 10A, 8000 Aarhus C,
Denmark
Email: [email protected]
2
which enables tri-planar motion. Unfortunately, the
unique motion of the thumb CMC joint renders it vulnerable to attrition of the articular surfaces leading to
osteoarthritis (OA), especially in postmenopausal
women, and to a lesser degree, in men (Bettinger
et al., 2001; Eaton and Littler, 1969; 1973). Depending
on the progression of the disease, symptoms range
from discomfort to the complete inability to manipulate objects (Cerveri et al., 2010; Xu et al., 1998).
The exact etiology of thumb CMC joint OA is disputed, but articular cartilage degeneration has been
attributed to the high loads transmitted across the
joint and to specific articular surface topography and
mechanics (Chao and An, 1978; Maxian et al., 1995). A
common rationalization is that ligament laxity and
incompetence lead to subluxation of the thumb metacarpal on the trapezium and areas of high contact
stress occur. This results in cartilage erosion, further
subluxation, and OA (Bettinger et al., 2001; Momose
et al., 1999).
The osteoarthritic changes of the thumb CMC
joint seen on radiographs are classified using the
four stage Eaton–Glickel classification of 1987 by
evaluation of osteophytes, joint space narrowing,
osteosclerosis, subluxation, and cystic changes in
the thumb CMC joint. In Eaton stage IV, there are
similar destructive changes in the scaphotrapezial
joint (Eaton and Glickel, 1987). Early joint degeneration (stage I) may be treated by conservative means
such as immobilization or steroid injections. In the
later stages (II, III, and IV) surgery may be required,
and commonly used methods include trapezectomy,
tendon interposition arthroplasty, and hemi and
total joint replacement surgery (Gangopadhyay
et al., 2012; Giddins, 2012; Johnston et al., 2011;
Kaszap et al., 2012; Klahn et al., 2012; Maru et al.,
2012; Poulter and Davis, 2011; Regnard, 2006;
Ulrich-Vinther et al., 2008).
Measurements of degenerative changes in the
CMC joint using plain radiographs are complicated by
errors due to over projection. The purpose of this
study is to gain a better understanding of the changes
that occur in the thumb CMC joint by quantitatively
comparing geometrical measurements in CT scans of
healthy and pathological thumb CMC joints.
Methods
All subjects underwent investigation including a CT scan
as part of routine clinical assessment. Thus, no extra
investigations were performed over and above our norm,
and so no consent was required. In accordance with local
ethical guidelines, no ethical approval was required for
this study because it was a medical database research
The Journal of Hand Surgery (Eur) 0(0)
project. Permission from the data protection agency was
obtained (Jr. Nr. 1-16-02-67-12).
Patient characteristics
We reviewed CT scans of patients with thumb CMC
joint OA and those with normal thumb CMC joints.
The osteoarthritic dataset of CT scans were
obtained at Holstebro Regional Hospital and Herning
Regional Hospital, Denmark. All subjects experienced
thumb pain and dysfunction. They were all graded for
thumb CMC OA by an experienced hand surgeon
(T.B.H.) using the Eaton–Glickel classification based
on standard radiographs (Hansen et al., 2012).
The control dataset of CT scans were obtained as
part of standard patient care, to assess both the
symptomatic and contralateral healthy wrist as a reference. These scans were collected over 2 years at
the Academic Medical Center, The Netherlands. An
experienced radiologist assessed all scans over this
period and excluded all wrists showing pathological
changes.
Image acquisition
The osteoarthritic dataset images were acquired
using 3 CT scanners: LightSpeed VCT; LightSpeed16
scanner (GE Medical Systems, Milwaukee, Wisconsin,
USA); and Brilliance 64 (Philips Medical Systems,
Best, The Netherlands). Control dataset images were
acquired using an Mx8000 Quad CT scanner (Philips
Medical Systems).
The acquisition parameters for all scans were similar. Representative acquisition parameters were collimation 2 x 0.5 mm, tube voltage 120 kV, effective
dose mAs 75, pitch 0.875, and slice thickness 0.6 mm;
scans were made in ‘ultra high resolution’ mode (i.e.,
small focal spot size). Tomographic reconstructions
were made with a high resolution kernel, field of view
of 150 mm, slice increment of 0.3 mm, and matrix of
512 x 512 pixels.
For all scans, in both the OA and control datasets,
the voxel (three-dimensional pixel) sizes differed, but
all deviated < 10% from a size of 0.3 x 0.3 x 0.3 mm3.
All scans were resampled to isotropic voxels of 0.3
mm using linear interpolation to simplify postprocessing. All analyses and measurements were
performed using MATLAB R2010a (MathWorks, Inc.,
Natick, Massachusetts, USA).
Semi-automatic segmentation
An initial, rough segmentation of the carpal bones was
obtained using simple thresholding and confirmed
3
Raedt et al.
with manual identification of each bone in MATLAB
R2010a (MathWorks, Inc.). Where thresholding did not
separate the bones, a manual correction was applied
to the thresholding result by manually annotating the
bone surface. This initial segmentation was obtained
and refined using a level set segmentation method
(Caselles et al., 1997; Sethian, 1999). Subsequently, a
marching cubes method was used to extract a triangulated bone surface from the segmentation results
(Lorensen and Cline, 1987). This method was previously found to be reproducible with a precision in the
order of a tenth of the voxel size for different CT scans
of the same carpal bones, while the accuracy was
judged visually to be in the order of the voxel size (van
de Giessen et al., 2009). An example of a fully segmented wrist can be seen in Figure 1.
Orientation and registration
Several of the measurements described in the following sections are orientation dependent. Therefore, all
wrists were mirrored to resemble a left wrist.
Furthermore, all scaphoids, trapezia, and metacarpals were aligned to a reference using a point cloud
registration method (Myronenko and Song, 2010).
This resulted in the scaphoids and trapezia of all the
wrists being aligned and scaled in order to remove
the effect of the difference in wrist size between subjects. This ensures that the difference in the metacarpal position between all subjects can be compared
quantitatively.
Measurements
In order to perform the analyses, it was necessary to
find the joint space of the thumb CMC joint and three
anatomical landmark points for each wrist. The
thumb CMC joint space was found first and then used
to derive the intercondylar point on the metacarpal
and reference point on the trapezium. A third reference point for the scaphoid was automatically determined, as described below.
Thumb CMC joint space identification
For each wrist, the joint space was defined by finding
the joint surface between the trapezium and metacarpal using a method described by van de Giessen
et al. (2009). The detection of the joint surface is
restricted by two conditions: (1) the angle between the
normal vectors of the adjacent bone surfaces should
deviate less than a predetermined angle and (2) the
maximum distance between adjacent bone surfaces
should be less than a predetermined distance. The
first condition ensures that the surface is only found
between bone surfaces facing each other; the second
condition prevents false detections of oppositely oriented, but non-adjacent surfaces. These settings only
affected the outer edge of the surface and were
experimentally determined to produce consistent
joint space surfaces for all included wrists (van de
Giessen et al., 2009).
Reference points
Three reference points were identified in each wrist:
the intercondylar point on the metacarpal, a reference point on the trapezium, and a reference point for
the scaphoid (Figure 2).
The intercondylar point was defined as the point
between the rounded protuberances at the base of
the metacarpal. Using the previously detected surface between the metacarpal and trapezium, reference points can be approximated by projecting the
centre of gravity of the thumb CMC joint surface onto
the metacarpal and trapezium, thus obtaining the
intercondylar point on the metacarpal and reference
point on the trapezium. The centre of gravity of each
scaphoid was calculated and used as the final reference point for the scaphoid. Using these reference
points, the following three measurements were
made: subluxation of the metacarpal with respect to
the trapezium; scaphoid-metacarpal height; and joint
space width.
Subluxation
Figure 1. An example of a left wrist with the scaphoid, trapezium, and metacarpal I labelled. The trapeziometacarpal
joint is located between the trapezium and metacarpal I.
Subluxation of the metacarpal was measured using
the location of the intercondylar point on the metacarpal in the three dimensional space of the aligned
wrists. Using the OA and control groups as two
4
The Journal of Hand Surgery (Eur) 0(0)
The joint space width is measured by calculating the
distance between the intercondylar point and trapezium reference point.
Data analysis
Figure 2. The points marked on the bones are the (A)
intercondylar point, (B) trapezium reference point, and (C)
scaphoid reference point.
separate classes, a linear discriminant analysis (LDA)
was performed to obtain the optimal axis along which
to measure the subluxation (Fisher, 1936). In this study,
LDA was used to find an axis along which the difference
in metacarpal position between the two groups could
be measured. This difference measured along the axis
is the subluxation of the metacarpal with respect to the
trapezium. Furthermore, the direction of the calculated axis indicates the direction of subluxation.
The data were split randomly using 20% of the
intercondylar metacarpal points for each group to
calculate the direction of the axis. The remaining 80%
of the two groups was used to quantify the accuracy of
the calculated direction. To verify that the axis was
calculated consistently, this procedure was repeated
100 times with the complete dataset.
Scaphoid-metacarpal height
The scaphoid-metacarpal height is defined as the distance between the scaphoid reference point and the
metacarpal intercondylar point (Figure 2). This height
represents a new measure to quantify the change in
the joint and we believe this may be correlated to joint
laxity.
Joint space width
The joint space width refers to the joint gap between the
articulating surfaces of the metacarpal and trapezium.
For each measurement three statistical analyses
were performed. The first analysis compared the control and OA groups. The second analysis compared
women to men within the OA group. The final analysis
compared OA stage II against stage III. Stages I and IV
were excluded due to insufficient numbers.
The resulting measurement data were tested for normality using the Shapiro–Wilk test in addition to skewness and kurtosis tests. Normally distributed data with
equal variances were tested with a t-test (two-group
mean-comparison test). Equality of variances were
tested with an F-test (two-group variance-comparison
test). Non-normal data were tested with the nonparametric Wilcoxon rank-sum test. The 95% confidence
intervals for non-normal data were estimated using
bias-corrected and accelerated bootstrapping from 4000
replicates. Significance levels were set at p < 0.05.
Results
The OA group consisted of 45 CT scans of patients (11
males, 34 females) with a mean age of 58 (40−72)
years. The control group consisted of 45 CT scans of
patients with a mean age of 38 (18−62) years. In the
OA group, there were one stage I, 18 stage II, 25 stage
III, and one stage IV on the Eaton–Glickel classification. All distances were measured in the coordinate
system of the reference subject.
Subluxation
The difference in the osteoarthritic group thumb metacarpal subluxation was found to be a mean of 2.45
mm (95% CI 2.06–2.77, p ≤ 0.001) compared with control (Figure 3). There was a significantly higher mean
difference of 0.74 mm (95% CI 0.18–1.31, p = 0.011) for
men compared with women (Figure 4). When stage II
and III cases were compared, there was a nonsignificant tendency to increased subluxation in stage
III of 0.37 mm (95% CI −0.10 to 0.84, p = 0.123). The
direction of subluxation determined by LDA was
radial-dorsal, as illustrated in Figure 5.
Scaphoid-metacarpal height
A decrease in scaphoid-metacarpal height of 1.65
mm (95% CI 1.20–2.10, p ≤ 0.001) was measured in
the osteoarthritic group compared with the control
5
Raedt et al.
Figure 3. Subluxation for control and OA groups. A significant difference can be seen between the control and osteoarthritic group.
Figure 4. Subluxation per sex for the OA group. Shows that males have a larger subluxation than females.
group (Figure 6). No statistically significant difference
in scaphoid-metacarpal height was found between
men and women.
The mean decrease in scaphoid-metacarpal
height was 0.41 mm (95% CI −0.13 to 0.86, p = 0.065)
more in stage III cases compared with stage II, which
might be clinically relevant, but was not statistically
significant.
Joint space width
A decrease in joint space width of 0.53 mm (95% CI
0.41–0.65, p ≤ 0.001) was measured in the
6
The Journal of Hand Surgery (Eur) 0(0)
Table 1. Subluxation, scaphoid-metacarpal height, and joint space width measurements per compared groups
Mean,
mm
Group
Subluxation
Scaphoid-metacarpal height
Joint space width
1.3
24.92
1
−1.90 to 2.31
21.76 to 27.23
0.54 to 1.64
−1.79
23.21
0.56
−2.59 to −0.60
20.87 to 24.36
0.14 to 1.27
−1.23
23.27
0.46
–2.73 to 0.83
21.51 to 24.65
0.09 to 1.44
Mean difference
(SD), mm
−2.73 to 1.25
20.87 to 26.00
0.01 to 1.44
−1.04
23.29
0.43
−2.73 to 1.25
21.51 to 26.00
0.01 to 1.44
2.45 (0.18)
1.65 (0.23)
0.53 (0.06)
−2.55 to −0.11
21.66 to 24.30
0.01 to 1.27
< 0.001*
< 0.001*
< 0.001*
0.74 (0.28)
0.08 (0.34)
−0.13 (0.10)
Stage III
−1.41
23.1
0.41
p value
Female
Stage II
−1.04
23.5
0.51
Range, mm
OA
Male
OA stage
Subluxation
Scaphoid-metacarpal height
Joint space width
Mean,
mm
Control
Sex
Subluxation
Scaphoid-metacarpal height
Joint space width
Range, mm
0.011*
0.862
0.086
0.37 (0.23)
0.41 (0.25)
0.10 (0.09)
0.123
0.065
0.14
*Statistically significant difference.
Figure 5. All intercondylar points shown together with the trapezium and reference axis. Arrow indicates the observed
subluxation in the radial-dorsal direction.
osteoarthritic groups compared with the control group
(Figure 7). No statistically significant difference was
found between men and women. No significant difference was found between stages II and III. A complete
summary of the results are presented in Table 1.
Discussion
We investigated the mechanical and anatomical
changes in the OA thumb CMC joint using CT scans. In
this study, we confirm that subluxation occurs in the
radial-dorsal direction and can be quantified using
the described technique.
We believe that subluxation may be an important
factor during joint replacement surgery. A common
implant design used for TM joint replacements is based
on a ball-and-socket joint. However, this replacement
and similar designs approximate the thumb CMC joint
kinematics poorly because the original saddle joint has
a variable centre of rotation (Giddins, 2012; Johnston
et al., 2011; Kaszap et al., 2012; Klahn et al., 2012; Maru
et al., 2012). The normal TM joint axis of rotation for
flexion-extension is located in the trapezium, while the
axis of rotation for abduction-adduction is located in the
metacarpal (Cerveri et al., 2010; Hollister et al., 1992;
Imaeda et al., 1996; Santos and Valero-Cuevas, 2006;
Valero-Cuevas et al., 2003). As these axes do not intersect, it is not possible to replace them with a single centre of rotation without changing the thumb CMC
kinematics. Furthermore, the design of ball-and-socket
Raedt et al.
7
Figure 6. Height from the scaphoid reference point to the intercondylar point on the metacarpal I. A decrease in height is
seen for the OA group.
Figure 7. Joint space width between the trapezium and metacarpal I. The width decrease represents the joint space narrowing characteristic of OA.
joints leads to an extra horizontal stress on the trapezium component.
These changes may contribute to the high failure
rates of ball-and-socket implants (Alnot and Saint
Laurent, 1985; Giddins, 2012; Hansen and Snerum,
2008; Hansen and Vainorius, 2008; Hernandez-Cortes
et al., 2012; Johnston et al., 2011; Kaszap et al., 2012;
Klahn et al., 2012; Maru et al., 2012; Pérez-Ubeda
et al., 2003; van Cappelle et al., 1999; Wachtl et al.,
1998). Therefore, it appears important that subluxation is recognized and that the centre of rotation is
restored as far as possible during prosthetic surgery.
8
Failure to correct the subluxation may lead to reduced
joint motion and increased horizontal forces at the
radial-dorsal rim of the cup and bone implant interfaces. This further aggravates the shortcomings of
the ball-and-socket design and ultimately leads to an
increased risk of implant failure. The results of this
study may help improve the design of thumb CMC
joint arthroplasties.
An important finding was that no significant difference in subluxation was found between stages II and
III. This shows that subluxation is probably not a good
criterion to differentiate between stages II and III. In
addition, the degree of subluxation is not included in
the Eaton–Glickel classification as originally described
and often, when considering subluxation, stage II and
III are seen as one group (Hansen et al., 2012).
Problems with intra-rater reliability with the Eaton–
Glickel classification may also bias this analysis. Due
to problems with the inter-rater reliability of the
Eaton–Glickel classification (Dela Rosa et al., 2004;
Kubik and Lubahn, 2002), we chose to have assessments of the radiographs performed by a single experienced hand surgeon (T.B.H.).
In this study all wrists were scaled in order to
remove the effect of wrist size. This is especially
important because of the differences between male
and female wrists. Despite the scaling performed, a
statistically significant difference was found in subluxation between females and males. The results
show that males experience a greater degree of subluxation for a similar stage of OA compared with
females. A plausible explanation may be that males
experience larger subluxation due to the difference in
wrist size.
The study also confirms what is widely known: that
a decrease in trapezium-metacarpal joint space width
occurs with OA due to both a reduction in cartilage
thickness and loss of bone stock. The reduction in joint
space width increased with progressing OA. Similarly,
we found that the scaphoid-metacarpal height was
significantly decreased in the OA group. We suspect
that a decrease in the scaphoid-metacarpal height is
due to a combination of subluxation, ligament laxity,
and loss of cartilage and even trapezium bone.
In advanced OA, the method to detect the bone
edges failed to separate adjacent bones because the
joint space was extremely narrow or missing (boneon-bone contact). This is a limitation of the method,
but was remedied by applying a manual correction to
the initial labelling.
Possible future research includes developing methods to perform measurements directly on CT images
without the need for segmenting the bone contours.
This would limit the processing time and allow for
The Journal of Hand Surgery (Eur) 0(0)
real-time use in a clinical setting. This could be implemented using active shape models, as described by
(Cootes et al., 1994). The resulting model may further
improve our understanding of changes in the joint in
patients with OA. In the future it would also be interesting to develop a method to measure the degree of
restoration of joint rotation after total joint replacement and relate this to implant success and failure.
In conclusion, this study helps the understanding
of the changes in joint anatomy in TM osteoarthritis
and may be useful for the design of total joint prosthesis and refinement of surgical technique.
Acknowledgements
We would like to thank Mario Maas for providing the control
wrist database used in this project. Further, our thoughts
and appreciation go to Kristian Larsen for his help with the
statistical analysis.
Conflict of interests
None declared.
Funding
This research received no specific grant from any funding
agency in the public, commercial, or not-for-profit sectors.
References
Alnot JY, Saint Laurent Y. Total trapeziometacarpal arthroplasty. Report on seventeen cases of degenerative
arthritis of the trapeziometacarpal joint. Ann Chir Main.
1985, 4: 11-21.
Bettinger PC, Linscheid RL, Cooney WP, An KN. Trapezial
tilt: a radiographic correlation with advanced trapeziometacarpal joint arthritis. J Hand Surg Am. 2001, 26:
692-7.
Caselles V, Kimmel R, Sapiro G. Geodesic active contours.
Int J Comput Vision. 1997, 22: 61-79.
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: 387-96.
Chao EY, An KN. Determination of internal forces in human
hand. J Eng Mech Div. 1978, 104: 255-72.
Cootes T, Hill A, Taylor C, Haslam J. Use of active shape
models for locating structures in medical images. Image
Vision Comput. 1994, 12: 355-65.
Dela Rosa TL, Vance MC, Stern PJ. Radiographic optimization of the Eaton classification. J Hand Surg Br. 2004,
29: 173-7.
Eaton RG, Littler JW. A study of the basal joint of the thumb.
Treatment of its disabilities by fusion. J Bone Joint Surg
Am. 1969, 51: 661-8.
Eaton RG, Littler JW. Ligament reconstruction for the painful thumb carpometacarpal joint. J Bone Joint Surg Am.
1973, 55: 1655-66.
Raedt et al.
Eaton RG, Glickel SZ. Trapeziometacarpal osteoarthritis.
Staging as a rationale for treatment. Hand Clin. 1987,
3: 455-71.
Fisher RA. The use of multiple measurements in taxonomic
problems. Ann Eugenic. 1936, 7: 179-88.
Gangopadhyay S, McKenna H, Burke FD, Davis TRC. Five- to
18-year follow-up for treatment of trapeziometacarpal
osteoarthritis: a prospective comparison of excision,
tendon interposition, and ligament reconstruction and
tendon interposition. J Hand Surg Am. 2012, 37: 411-7.
Giddins G. Editorial. J Hand Surg Eur Vol. 2012, 37: 603-4.
Hansen TB, Snerum L. Elektra trapeziometacarpal prosthesis for treatment of osteoarthrosis of the basal joint
of the thumb. Scand J Plast Reconstr Surg Hand Surg.
2008, 42: 316-9.
Hansen TB, Vainorius D. High loosening rate of the Moje
Acamo prosthesis for treating osteoarthritis of the trapeziometacarpal joint. J Hand Surg Eur Vol. 2008, 33:
571-4.
Hansen TB, Sorensen OG, Kirkeby L, Homilius M, Amstrup AL.
Computed tomography improves intra-observer reliability, but not the inter-observer reliability of the EatonGlickel classification. J Hand Surg Eur Vol. 2012 (epub).
Hernandez-Cortes P, Pajares-Lopez M, Robles-Molina MJ,
Gomez-Sanchez R, Toledo-Romero MA, De TorresUrrea J. Two-year outcomes of Elektra prosthesis for
trapeziometacarpal osteoarthritis: a longitudinal cohort
study. J Hand Surg Eur Vol. 2012, 37: 130-7.
Hollister A, Buford WL, Myers LM, Giurintano DJ, Novick A.
The axes of rotation of the thumb carpometacarpal joint.
J Orthop Res. 1992, 10: 454-60.
Imaeda T, Cooney WP, Niebur GL, Linscheid RL, An KN.
Kinematics of the trapeziometacarpal joint: a biomechanical analysis comparing tendon interposition
arthroplasty and total-joint arthroplasty. J Hand Surg
Am. 1996, 21: 544-53.
Johnston P, Getgood A, Larson D, Chojnowski AJ,
Chakrabarti AJ, Chapman PG. De la Caffiniere thumb
trapeziometacarpal joint arthroplasty: 16-26 year follow-up. J Hand Surg Eur 2011, 37: 621-4.
Kaszap B, Daecke W, Jung M. High frequency failure of the
Moje thumb carpometacarpal joint arthroplasty. J Hand
Surg Eur Vol. 2012, 37: 610-6.
Klahn A, Nygaard M, Gvozdenovic R, Boeckstyns ME.
Elektra prosthesis for trapeziometacarpal osteoarthritis: a follow-up of 39 consecutive cases. J Hand Surg Eur
Vol. 2012, 37: 605-9.
Kubik NJ, Lubahn JD. Intrarater and interrater reliability
of the Eaton classification of basal joint arthritis. J Hand
Surg Am. 2002, 27: 882-5.
Lorensen WE, Cline HE. Marching cubes: A high resolution
3D surface construction algorithm. Comp Graph. 1987,
21: 163-9.
Maru M, Jettoo P, Tourret L, Jones M, Irwin L. Thumb
carpometacarpal osteoarthritis: trapeziectomy versus
9
pyrocarbon interposition implant (Pi2) arthroplasty. J
Hand Surg Eur Vol. 2012, 37: 617-20.
Maxian TA, Brown TD, Weinstein SL. Chronic stress tolerance levels for human articular cartilage: two nonuniform contact models applied to long-term follow-up of
CDH. J Biomech. 1995, 28: 159-66.
Momose T, Nakatsuchi Y, Saitoh S. Contact area of the trapeziometacarpal joint. J Hand Surg Am. 1999, 24: 491-5.
Myronenko A, Song X. Point set registration: coherent point
drift. IEEE T Pattern Anal. 2010, 32: 2262-75.
Pérez-Ubeda MJ, García-López A, Marco Martinez F,
Junyent Vilanova E, Molina Martos M, López-Duran
Stern L. Results of the cemented SR trapeziometacarpal
prosthesis in the treatment of thumb carpometacarpal
osteoarthritis. J Hand Surg Am. 2003, 28: 917-25.
Poulter RJ, Davis TRC. Management of hyperextension of
the metacarpophalangeal joint in association with trapeziometacarpal joint osteoarthritis. J Hand Surg Eur
Vol. 2011, 36: 280-4.
Regnard P-J. Electra trapezio metacarpal prosthesis:
results of the first 100 cases. J Hand Surg Br. 2006, 31:
621-8.
Santos VJ, Valero-Cuevas FJ. Reported anatomical variability naturally leads to multimodal distributions of
Denavit-Hartenberg parameters for the human thumb.
IEEE T Biomed Eng. 2006, 53: 155-63.
Sethian JA. Level set methods and fast marching methods:
evolving interfaces in computational geometry, fluid
mechanics, computer vision, and materials science.
Berkeley, Cambridge University Press, 1999.
Ulrich-Vinther M, Puggaard H, Lange B. Prospective 1-year
follow-up study comparing joint prosthesis with tendon interposition arthroplasty in treatment of trapeziometacarpal osteoarthritis. J Hand Surg Am. 2008, 33:
1369-77.
Valero-Cuevas FJ, Johanson ME, Towles JD. Towards a
realistic biomechanical model of the thumb: the choice
of kinematic description may be more critical than the
solution method or the variability/uncertainty of musculoskeletal parameters. J Biomech. 2003, 36: 1019-30.
van Cappelle HG, Elzenga P, Van Horn JR. Long-term
results and loosening analysis of de la Caffinière
replacements of the trapeziometacarpal joint. J Hand
Surg Am. 1999, 24: 476-82.
van de Giessen M, Streekstra G, Strackee S et al.
Constrained registration of the wrist joint. IEEE T Med
Imaging. 2009, 28: 1861-9.
Wachtl SW, Guggenheim PR, Sennwald GR. Cemented and
non-cemented replacements of the trapeziometacarpal
joint. J Bone Joint Surg Br. 1998, 80: 121-5.
Xu L, Strauch R, Ateshian G, Pawluk R, Mow V, Rosenwasser
M. Topography of the osteoarthritic thumb carpometacarpal joint and its variations with regard to gender, age,
site, and osteoarthritic stage. J Hand Surg Am. 1998, 23:
454-64.