Surg Radiol Anat DOI 10.1007/s00276-015-1418-7
ORIGINAL ARTICLE
Dimensions of the trapezium bone: a cadaver and CT study F. Loisel • S. Chapuy • P.-B. Rey • L. Obert B. Parratte • L. Tatu • D. Lepage
•
Received: 29 April 2014 / Accepted: 2 January 2015 Ó Springer-Verlag France 2015
Abstract Purpose The primary purpose of this study was to define the size of the trapezium bone through measurements on cadaver specimens and CT scans of living subjects. The secondary purpose of this study was to determine if any correlation existed between the size of the trapezium and local anatomical parameters. Methods The radio-ulnar length (L), dorsopalmar width (‘) and height (h) of the distal surface of the trapezium were measured by two independent observers on 20 cadaver specimens. The same measurements were carried out by two other observers on anonymized CT scans from 18 patients. The inter- and intra-observer agreement was determined using the intraclass correlation coefficient. F. Loisel (&) S. Chapuy B. Parratte L. Tatu D. Lepage Laboratoire d’Anatomie, UFR SMP, 20 rue Ambroise Pare´, Besanc¸on, France e-mail:
[email protected];
[email protected] F. Loisel P.-B. Rey L. Obert D. Lepage Service d’orthope´die, de traumatologie, de chirurgie plastique, reconstructrice et assistance main, EA 4268 innovation, imagerie, inge´nierie et intervention en sante´ I4S-IFR 133 Inserm, CHU Jean-Minjoz, Universite´ de Franche-Comte´, Besanc¸on, France S. Chapuy Service de radiologie Visce´rale, CHU Jean-Minjoz, Universite´ de Franche-Comte´, Besanc¸on, France B. Parratte Service de me´decine physique et de re´adaptation, CHU JeanMinjoz, Universite´ de Franche-Comte´, Besanc¸on, France L. Tatu Service d’Explorations et Pathologies Neuro-musculaires, CHU Jean-Minjoz, Universite´ de Franche-Comte´, Besanc¸on, France
Results In the cadavers, the mean length, width and height of the trapezium were 22.8, 15.5 and 15.2 mm, respectively. On the CT scans, these same dimensions were 19.2, 11.4 and 11.6 mm. Inter-observer agreement was statistically significant in both parts of the study. Discussion The dimensions of the trapezium bone were about 3.33 mm larger in cadavers than on CT scans. These differences can be explained partially by a systematic under-sizing error on the CT scans and the fact that the cartilage layer cannot be directly visualized. Conclusion This study was able to define the dimensions of the trapezium bone. It may be possible to predict the trapezium height from the length of the forearm or the width of the radial epiphysis. Our data can be used to adjust the size of trapezium implants to the dimensions of the patient’s bone. Keywords Trapezium bone Wrist anatomy Trapezial implant CT scan
Introduction The size of the trapezium bone must be taken into consideration when implanting a trapeziometacarpal prosthesis. But this joint replacement procedure is often performed based on very small trapezium bones [4]. There is no anatomical information about the dimensions of the trapezium in its native environment. We report a two-part study consisting of cadaver and CT scan evaluations performed to better assess the dimensions of the trapezium bone, with the underlying goal of improving conformity between trapezium implants and the trapezium bone itself.
123
Surg Radiol Anat
Materials and methods Cadaver study In the cadaver study, 20 trapezium bones (10 right and 10 left) were dissected. These dissections were carried out in the Anatomy Laboratory of one of the author’s affiliated university. The cadavers had been fresh frozen in 8 cases and formalin fixed in 12. There were 11 female and 9 male subjects. One of the female subjects was excluded because the trapezium was too damaged for size measurements to be carried out. The average age at death was 83 years (range 63–94); the average height was 161 cm (range 150–176); the average forearm length was measured from the tip of the olecranon to the anterior distal palmar wrist crease. A-P radiographs were taken of the dissected wrists. The metacarpal surface (distal surface) of the trapezium was used as a reference for the various measurements. Two independent observers measured the radio-ulnar length (L), dorsopalmar width (‘), and height (h) of the trapezium, along with the width of the distal radial epiphysis (E) (Fig. 1). The radio-ulnar length was measured on half of the metacarpal surface from the radial limit to the ulnar limit of the articular surface. The dorsopalmar width was measured on half of the distal surface of the trapezium bone (metacarpal surface), from the palmar limit to the dorsal limit of the articular surface. The anatomical specimens and the associated wrist radiographs were
Fig. 1 Cadaver measurements. a Measurement of trapezium height (h) on volar aspect. b Measurement of length (L) and width (‘) on the metacarpal surface. c Measurement of the distal radial epiphysis width (E) on an A-P radiograph of the wrist
123
photographed with a Lumix DMC-FZ48Ò camera. On each photograph, a known gauge length was used to determine the L, ‘, h and E dimensions using the tools provided in the GIMP imaging software (version 2.8.4). Signs of osteoarthritis were noted and used to classify the samples into one of three groups: total arthritis if [50 % of the articular surface was damaged, partial arthritis if \50 % of the articular surface was damaged and no arthritis if the articular surface was undamaged. CT scan study The CT component of this study was performed on 18 anonymized wrists CT scans (Brilliance CT 64-channel, Philips, The Netherlands). These were taken from nine women and nine men having an average age of 71 years (range 55–86). The slices were less than 1 mm thick (range 0.55–1). The slices were chosen by the radiologist according to the CT indication. Two independent observers, who were not involved in the cadaver study, measured the radio-ulnar length (L), dorsopalmar width (‘), and height (h) of the trapezium, along with the width of the distal radial epiphysis (E) using open-source OsiriXÒ imaging software (Fig. 2). The software generated a 3D reconstruction of the trapezium, which made it possible to carry out the measurements in the same planes as the cadaver study. Each observer performed the measurements twice, with at least 15 days between measurement sessions. The presence of basal joint osteoarthritis was evaluated using Dell’s classification.
Fig. 2 CT scan measurements. a Measurement of the height (h) of the trapezium on CT scan. b Measurement of the length (L) and width (‘) of the metacarpal surface of the trapezium on a CT scan. c Measurement of the width of the radial epiphysis (E) on CT scan
Surg Radiol Anat
Statistical analysis The statistical analysis was performed with XLSTATÒ software (Addinsoft, New York, NY, USA). The interobserver agreement was determined using the intraclass correlation coefficient (ICC) for the cadaver and CT scan components of the study. For the CT portion of the study, the intra-observer agreement was also determined. Once the normality of the data set was confirmed, Student’s t test was performed to compare the results between men and women. Pearson’s correlation test was used to determine whether a relationship existed between the various measurements.
Results In the cadavers, the mean length (L), width (‘) and height (h) of the trapezium were 22.8, 15.5 and 15.2 mm, respectively. The average forearm length was 25.9 cm (range 22–31). On the CT scan images, the mean length (L), width (‘) and height (h) of the trapezium were 19.2, 11.4 and 11.6 mm, respectively. The average radial epiphysis width (E) was 29.8 mm on the cadaver specimens and 32.3 mm on the CT scans. In both studies, the mean dimensions of the trapezium differed significantly between genders, except for the length and width in the cadaver study and for the width of the trapezium in CT study (Tables 1, 2). The bone dimensions on the cadaver specimens were 3.33 mm larger on average than on the CT scan images (Table 3). In the cadaver study, there was a significant correlation between the trapezium height and length of the forearm (P = 0.030 in men and P = 0.012 in women) (Figs. 3, 4). The length and width of the trapezium were not correlated with the forearm length, radial epiphysis width or subject height. Sixteen of trapezium bones had signs of osteoarthritis: ten total and six partial. Three of the trapezium bones had no signs of osteoarthritis. In the CT scan study, there was a significant correlation (P \ 0.05) between the width of the radial epiphysis and the various dimensions of the trapezium bone. There were five cases of basal joint
arthritis in the CT scans: two were grade I, two were grade II and one was grade IV. There was a significant correlation between the measurements made by the various observers in the cadaver study (ICC [80 %). The intra-observer ICC was greater than 90 % for all CT measurements, except for width (ICC of 79 %). The inter-observer ICC was 96 % for the radial epiphysis width, 96 % for the trapezium length, 77 % for its width and 89 % for its height.
Discussion Although many studies have described the dynamic aspects of carpal bones and their volumes, few studies have looked into the morphometry of the trapezium bone [1, 16]. Published studies range in scope from evaluations of the radius of curvature on the metacarpal surface, bone architecture of the trapezium, vascularity, functional anatomy to the ligament environment [6, 8, 11, 13, 14]. One of the strengths of the current study is that physiological dimensions of the trapezium were determined in the bone’s anatomical environment. The two main limitations of this study revolve around the subject age and the method used for preserving the cadavers, which can overestimate the dimensions of the trapezium. By measuring the trapezium size on both cadavers and CT scans, we found the cadaver-based trapezium measurements, which had a mean trapezium length of 22.8 mm, a width of 15.5 mm and a height of 15.2 mm, to be about 3.33 mm greater than those on the CT scanner. What can explain these differences? There could have been a systematic under-sizing error of the trapezium on the CT scan. However, various studies have shown that this error is minimal, ranging from less than 1 mm (maximum 0.47 mm) [15, 21] to about 1 mm [10, 20]. The explanation is thought to be related to the CT technology itself, as the cartilage layer cannot be directly visualized on CT scans. Several studies have defined the cartilage thickness of certain joints, including the wrist joint [7, 17]. Based on 104 trapeziometacarpal joints analyzed by stereophotogrammetry and histology, the trapezium cartilage was 0.8 ± 0.2 mm thick and the cartilage
Table 1 Trapezium bone dimensions by gender in the cadaver specimens Forearm length (mm)
Radial epiphysis width (mm)
Trapezium length (mm)
Trapezium width (mm)
Trapezium height (mm)
Men (n = 9)
278 ± 17
31.23 ± 1.2
22.95 ± 1.8
16.32 ± 1.5
16.07 ± 1.8
Women (n = 10)
244 ± 11
28.29 ± 1.0
22.63 ± 2.1
14.84 ± 1.8
14.4 ± 1.3
Men/women differences
P \ 0.0001
P \ 0.0001
NS
NS
P \ 0.0001
123
Surg Radiol Anat Table 2 Trapezium bone dimensions by gender in the CT scan study Radial epiphysis width (mm)
Trapezium length (mm)
Trapezium width (mm)
Trapezium height (mm)
Men (n = 9)
34.94 ± 1.8
20.13 ± 0.7
11.70 ± 0.5
12.57 ± 0.6
Women (n = 9)
29.62 ± 1.8
18.26 ± 1.8
11.1 ± 0.8
10.62 ± 1.3
Men/women differences
P = 0.007
P = 0.01
P = 0.07
P = 0.005
Table 3 Measured differences between the cadaver and CT scan study
Cadaver study (mm)
CT study (mm)
Difference (mm)
Length (mm) Men
22.95 ± 1.8
20.13 ± 0.7
2.82
Women
22.63 ± 2.1
18.26 ± 1.8
4.37
Men
16.32 ± 1.5
11.70 ± 0.5
4.62
Women
14.84 ± 1.8
11.1 ± 0.8
3.74
16.07 ± 1.8 14.40 ± 1.3
12.57 ± 0.6 10.62 ± 1.3
3.5 3.78
Width (mm)
Height (mm) Men Women
3.33
R²=0.567
Height male trapezium bone (mm)
Height female trapezium bone (mm)
Mean
21 19 17 15 13 11 9 7 22
23
24
25
26
27
Length female forearm (cm)
R²=0.512
25 23 21 19 17 15 13 11 9 7 24
25
26
27
28
29
30
31
32
Length male forearm (cm)
Measurement
Linear regression it
Measurement
Linear regression it
95% C.I. (observed)
95% C.I. (observed)
95% C.I. (mean)
95% C.I. (observed) )
Fig. 3 Linear regression of the trapezium height and forearm size in women
Fig. 4 Linear regression of the trapezium height and forearm size in men
on the metacarpal side was 0.7 ± 0.2 mm [12] thick. Similar results were found in a study comparing optical coherence tomography with 3D tomography. The trapeziometacarpal cartilage was between 0.6 and 1.0 mm thick on the CT scan, but 1.1 mm in histomorphometry [2]. However, the cartilage thickness only affects the height measurement; it cannot explain the observed length and width differences. A comparison of the results of the current study relative to those of published cadaver [1] and CT scan studies [3, 9,
16] is given in Tables 4 and 5. The difference between the findings of the current study on cadavers with the one by Bonnel et al. can be explained by differences in how the cadavers were stored and in how different landmarks were used for the measurements. Differences in landmarks for the trapezium measurements on the CT scans can also explain the differences observed. In the Patterson study, for example, the major axis of all the carpal bones was measured, but not the length, width or height. Similar observations apply to the Crisco et al. [3] study, where the goal
123
Surg Radiol Anat Table 4 Comparison of the current study’s cadaver-based results with published results Study Bonnel et al. (1990) [2] Current study (2014)
Number of specimens
Gender
Trapezium length (mm)
Trapezium width (mm)
Trapezium height (mm)
8
Male
15.8 ± 1.6
10.7 ± 0.5
4
Female
13.8 ± 0.7
10.5 ± 0.6
10.4 ± 0.3
9
Male
22.95 ± 1.8
16.32 ± 1.5
16.07 ± 1.8
Female
22.63 ± 2.1
14.84 ± 1.8
14.40 ± 1.3
10
11.8 ± 0.6
Table 5 Comparison of the current study’s CT-based results with published results Study
Number of subjects
Age
Basal joint arthritis
CT slice thickness (mm)
Patterson et al. (1995) [3]
35
17–89
Not recorded
1.5
Not recorded Not recorded
Crisco et al. (2005) [17]
Gender
Trapezium length (mm)
Trapezium width (mm)
Trapezium height (mm)
Male
25.91 ± 3.39 (measured on major axis)
Female
23.02 ± 2.09 (measured on major axis)
1
Male
25.4 ± 1.8
17.5 ± 1.8
16.1 ± 1.8
1
Female
21.8 ± 1.8
15.8 ± 1.5
13.1 ± 1.2
Gender ratio M/F = 1.7
16.9 ± 1.5
20.9 ± 2.7
18.7 ± 2.7
14
25.6
14
23.6
Hansen et al. (2011) [18]
31
57
31
2.5
Current study (2014)
18
71
5
0.67
was to compare carpal bone sizes between men and women based on volume reconstruction. The primary landmark was the center of inertia. Although the Crisco results were similar to our cadaver-based evaluation, no definitive conclusions should be drawn. Age also induces some variability; it was 71 years in the current study versus 57 in one study [9] and between 23 and 25 in another [3]. A selection bias was also at the source of the observed differences. Among the 18 CT scan performed, there were 5 cases of basal joint arthritis; in the Hansen study, 31 of the CT scans were performed in patients with Eaton grade 2 or 3 basal joint arthritis [5]. It appears that the distal radial epiphysis width and forearm length are correlated with the trapezium height. We also found significant differences between the trapezium size in men and women, as in other published studies [3, 19]. One of the applications of this study’s results is trapeziometacarpal osteoarthritis surgery. In the early stage of the disease, Kapandji and Heim propose performing a trapezium osteotomy if the trapezial slope is pathological (i.e., the angle between the longitudinal axis of the second metacarpal and the axis of the trapeziometacarpal joint is more than 130°). This technique is technically demanding and there is a possibility of damaging the trapezium
Male
20.13 ± 0,7
11.70 ± 0.5
12.57 ± 0.6
Female
18.26 ± 1,8
11.1 ± 0.8
10.62 ± 1,3
cartilage or tendons [18]. Knowledge of the average height of the trapezium and its other parameters could improve this procedure. Another potential application relates to trapeziometacarpal arthroplasty. There is a notable disparity in the size of various commercially available trapezium implants [4]. In 90 % of cases, these cup-shaped implants are defined by their diameter and depth (height). In the 20 available models, the diameter ranges from 5 to 12 mm (mean of 8.94 mm) and depth from 4 to 8.2 mm (mean of 5.97 mm). If these sizes are compared to the measured width (15.5 mm) and height (15.2 mm) of cadaver trapezium bones, the prosthetic trapezium implants are under-sized. These implants could be improved by making them bigger so they more closely match the size of the native trapezium. By increasing the implant size, the thickness of the polyethylene insert would also be increased and the implant wear reduced. However, other studies are needed to validate this hypothesis: one must bear in mind the scale of the anatomically studied region (difference on the order of millimeters); the risk of fracture during trapezial cup implantation cannot be ignored. Most manufacturers offer two types of implants by fixation type. Increasing the number of implants offered would provide a better match with the size of the patient’s trapezium. The implant size could then be adapted to the patient’s morphology and
123
Surg Radiol Anat
gender, which is currently the case with some total knee arthroplasty implants.
Conclusion This study was able to define the size of the trapezium bone. It seems possible to predict trapezium height from the size of the forearm or the width of the distal radial epiphysis. These data could be used as a basis for designing new implants for the treatment of basal joint arthritis, with a larger size and wider range of available sizes being the primary emphasis. These results would also allow surgical planning to take into consideration the patient’s gender and bone morphology when selecting the implant size. Additional studies with a larger number of subjects would be useful, especially for determining the degree to which the trapezium dimensions change due to osteoarthritis (wider distal surface, but reduced height). If the trapezium appears ‘‘small’’ on radiographs, a CT scan should be performed to better estimate the size of the trapezium and determine if it can accept a trapeziometacarpal prosthesis. Acknowledgments The authors would like to thank Hugues Grandin for the preparation of the cadavers and Joanne Archambault, PhD for the editorial assistance provided during the preparation of this manuscript. Conflict of interest of interest.
The authors declare that they have no conflict
References 1. Bonnel F, Bellan N (1990) L’articulation trape´zo-me´tacarpienne du pouce. In: Saffar P (ed) La Rhizarthrose. Expansion Scientifique Franc¸aise, Paris 2. Cernohorsky P, de Bruin DM, van Herk M, Bras J, Faber DJ et al (2012) In-situ imaging of articular cartilage of the first carpometacarpal joint using co-registered optical coherence tomography and computed tomography. J Biomed Opt 17:060501 3. Crisco JJ, Coburn JC, Moore DC, Upal MA (2005) Carpal bone size and scaling in men versus in women. J Hand Surg 30:35–42 4. Daunois O, Durand S, Gaujoux G, Me´o S, Strube F, Sassoon D (2011) Les cupules trape´ziennes. Chir Main 30:83–85 5. Eaton RG, Glickel SZ (1987) Trapeziometacarpal osteoarthritis. Staging as a rationale for treatment. Hand Clin 3:455–471 6. Edmunds JO (2011) Current concepts of the anatomy of the thumb trapeziometacarpal joint. J Hand Surg 36:170–182
123
7. Foumani M, Strackee SD, van de Giessen M, Jonges R, Blankevoort L, Streekstra GJ (2013) In-vivo dynamic and static threedimensional joint space distance maps for assessment of cartilage thickness in the radiocarpal joint. Clin Biomech Bristol Avon 28:151–156 8. Goubau JF, Benis S, Van Hoonacker P, Berghs B, Kerckhove D, Patonay L (2012) Vascularization of the trapeziometacarpal joint and its clinical importance: anatomical study. Chir Main 31:57–61 9. Hansen TB, Hengst D, Mortensen J, Amstrup AL (2011) Fixation of trapezial implants in a trapeziometacarpal total joint prosthesis tested in a model of porcine bone. J Plast Surg Hand Surg 45:263–266 10. Hildebolt CF, Vannier MW, Knapp RH (1990) Validation study of skull three-dimensional computerized tomography measurements. Am J Phys Anthropol 82:283–294 11. Humes D, Ja¨hnich H, Rehm A, Compson JP (2004) The osteology of the trapezium. J Hand Surg Edinb Scotl 29:42–45 12. Koff MF, Ugwonali OF, Strauch RJ, Rosenwasser MP, Ateshian GA, Mow VC (2003) Sequential wear patterns of the articular cartilage of the thumb carpometacarpal joint in osteoarthritis. J Hand Surg 28:597–604 13. Lee AT, Williams AA, Lee J, Cheng R, Lindsey DP, Ladd AL (2013) Trapezium trabecular morphology in carpometacarpal arthritis. J Hand Surg 38:309–315 14. Marzke MW, Tocheri MW, Marzke RF, Femiani JD (2012) Three-dimensional quantitative comparative analysis of trapezialmetacarpal joint surface curvatures in human populations. J Hand Surg 37:72–76 15. Oka K, Murase T, Moritomo H, Goto A, Sugamoto K, Yoshikawa H (2009) Accuracy analysis of three-dimensional bone surface models of the forearm constructed from multidetector computed tomography data. Int J Med Robot 5:452–457 16. Patterson RM, Elder KW, Viegas SF, Buford WL (1995) Carpal bone anatomy measured by computer analysis of three-dimensional reconstructions of computed tomography images. J Hand Surg 20:923–929 17. Pollock J, O’Toole RV, Nowicki SD, Eglseder WA (2013) Articular cartilage thickness at the distal radius: a cadaveric study. J Hand Surg 38:1477–1481 (discussion 1482–1483) 18. Ropars M, Siret P, Kaila R, Marin F, Belot N, Dre´ano T (2009) Anatomical and radiological assessment of trapezial osteotomy for trapezial dysplasia in early trapeziometacarpal joint arthritis. J Hand Surg Eur 34:264–267 19. Schuind FA, Linscheid RL, An KN, Chao EY (1992) A normal data base of posteroanterior roentgenographic measurements of the wrist. J Bone Joint Surg Am 74:1418–1429 20. Vannier MW (2000) Evaluation of 3D imaging. Crit Rev Diagn Imaging 41:315–378 21. Wang J, Ye M, Liu Z, Wang C (2009) Precision of cortical bone reconstruction based on 3D ct scans. Comput Med Imaging Graph 33:235–241