- Email: [email protected]
Ankle morphometry evaluated using a new semi-automated technique based on X-ray pictures
Clinical Biomechanics 20 (2005) 307–311 www.elsevier.com/locate/clinbiomech
Ankle morphometry evaluated using a new semi-automated technique based on X-ray pictures Rita Stagni a
a,*
, Alberto Leardini b, Andrea Ensini b, Angelo Cappello
a
Dipartimento di Electronica, Informatica e Sistemistica, Universita` degli Studi di Bologna, Viale Risorgimento 2, 40136 Bologna, Italy b Laboratorio Analisi del Movimento, Istituti Ortopedici Rizzoli, Via di Barbiano 1/10, 40136 Bologna, Italy Received 1 October 2004; accepted 15 November 2004
Abstract Background. The study of bone morphometry is important in anatomy, biomechanics and the design of joint arthroplasty, because the design of prostheses depends also on the knowledge of joint geometry, both for the replication of function and for the proper sizing of the components. Relevant information for the human ankle joint bones is particularly scarce. Methods. A large number of morphological measurements of the ankle joint were collected in 36 normal subjects by means of a new semi-automated radiographic measurement method based on standard frontal and sagittal X-ray projections. Lengths, heights and widths of the main parts of this articulation together with the radii of curvature of the talar dome and of the trochlea tali in the sagittal plane were worked out. Statistical analysis was performed on these measurements. Findings. A large spectrum for the values of these measures was found. Significant correlation was found among some of the measures, but not between any of these and the subject height. Interpretation. Bones at the ankle joint have similar shapes, i.e. the three dimensions are in direct proportion to each other. No one of the internal dimensions is correlated with the only one possible external measurement, i.e. the malleolar width. Sizes of the current total ankle replacement designs may differ considerably from real corresponding joint dimensions. 2004 Elsevier Ltd. All rights reserved. Keywords: Morphometry; X-rays; Talo-crural joint; Curvature radii; Prosthesis size; Automated measurements
1. Introduction Clinical success of all kinds of human joint arthroplasty depends also on available information on the morphology of the relevant bones. For example, the disappointing clinical results of current ankle implants have been primarily related to unsatisfactory prosthesis design. The design of this replacement relies on the understanding of natural mobility and stability (Leardini, 2001), and thus, on a thorough knowledge of joint geometry. Moreover, in designing human joint prostheses, also the geometry of the components needs atten*
Corresponding author. E-mail address: [email protected] (R. Stagni).
0268-0033/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2004.11.009
tion, particularly with regards to the different size dimensions. Although several studies for other human joints are reported in the literature (Erkman and Walker, 1974; Roberts et al., 1991; McCarthy et al., 1997), studies about the morphology of the bones forming the ankle or talo-crural joint are scarce. Only a few studies have reported on ankle joint geometrical measurements (Kempson et al., 1975; Mariani and Patella, 1977; Fessy et al., 1997). Although large samples have been analysed by these authors (41, 100 and 50 subjects, respectively), the reliability of the measurements has never been reported, and even magnification not taken into account (Fessy et al., 1997). Accuracy of the method adopted to perform the morphological analysis is a fundamental issue. The method
308
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
sphere (20 mm diameter, 0.01 mm tolerance) in front of the ankle in the lateral projection, and over the lateral malleolus in the antero-posterior projection. This was necessary to calculate X-ray magnification in each patient exactly at the level of the plane under analysis (Stagni et al., 2004). The main general characteristics of these subjects are reported in Table 1. The X-ray pictures were taken in un-loaded conditions when the subject was lying supine on the radiological bed, the foot was in neutral position with respect to the shank and approximately half of the shank and of the foot were visible. Morphological measurements were taken for each pair of X-rays using a software package designed for the purpose. Relevant accuracy was assessed to be, in a previous specifically addressed study (Stagni et al., 2004), within 1 mm and repeatability was not affected by operatorÕs skill. This software allows a series of measurements to be taken on the bone silhouettes, schematically shown in Fig. 1. The measurements taken from the sagittal projection (Fig. 1(a)) were
must allow large groups of subjects to be measured, with minimal invasiveness in case of in vivo studies, and reliable data to be produced in order to perform valuable statistical analysis. Although standard radiograms are routinely used to perform morphological measurements (Mensch and Amstutz, 1975; Mariani and Patella, 1977; Elias et al., 1990; Fessy et al., 1997), the relevant reliability was found to be significantly dependent on the operatorÕs skill (Brage et al., 1997). Although recent studies (Nakai et al., 2000; Tanaka et al., 2000) still measured joint morphometry from radiograms directly by using protractors or goniometers, current recommendations in this respect (Coughlin et al., 2002) pointed out the necessity of awareness. Others (Prakash et al., 2001) already demonstrated the important improvement in terms of accuracy and repeatability of the measurements performed using automatic procedures. A new semi-automatic method based on standard radiograms (Stagni et al., 2004) has been recently proposed and validated for the measurement of a large number of morphometric variables at the human ankle joint. This method was assessed to be accurate, repeatable and little dependant on operatorÕs skill. A relevant statistical analysis of these variables is of fundamental importance for the design and sizing of ankle prostheses and of the relevant surgical instruments. The purpose of this study was to report on a set of morphometrical measurements at the human ankle joint obtained from a large sample of normal subjects by using the novel method (Stagni et al., 2004).
• Tibial arc length (TiAL): length of the segment connecting the most posterior (B) and the most anterior (A) points of the tibial mortise. • Sagittal radius of the tibial mortise (SRTi): radius of the circumference fitting the points of the tibial mortise profile. • Antero-posterior gap (APG): distance along the tibial longitudinal axis between A and B. • Antero-posterior inclination angle of the tibial mortise (APA): angle between the segment AB and the tibial antero-posterior axis. • Maximal tibial thickness (MTiTh): distance between the most anterior point of the tibial anterior profile (C) and the corresponding point (D) along the tibial antero-posterior axis on the posterior tibial profile. • Distance of level of MTiTh from the anterior limit of the mortise (MDA): distance along the tibial longitudinal axis between A and C. • Distance of level of MTiTh from the vertex of the mortise (MDV): distance along the tibial longitudinal axis between the vertex of the tibial mortise V, defined as the intersection between the tibial longitudinal axis and the tibial mortise profile, and C.
2. Methods Lateral and frontal X-rays pictures of the ankle joint were acquired in 36 subjects (13 females, 23 males). These patients were selected among those reporting with an ankle sprain or trauma at the Emergency Room of the authorÕs Institute in the period November 2000–January 2001. Among these, when standard diagnostic imaging procedure was ordered, only those to which bony, joint and syndesmotic integrity was assessed before and after the standard imaging procedure were included in the present analysis. These consented to modify the standard procedure by simply sticking a steel
Table 1 Mean value, standard deviation, maximum and minimum values of age (years), weight (kg) and height (m) of the group of 23 men and 13 women All (n = 36)
Mean SD Max Min
Male (n = 23)
Female (n = 13)
Age (years)
Weight (kg)
Height (m)
Age (years)
Weight (kg)
Height (m)
Age (years)
Weight (kg)
Height (m)
38 17 76 15
71.9 11.1 91.0 52.0
1.72 0.09 1.86 1.55
33 17 76 15
74.0 10.4 91.0 52.0
1.76 0.08 1.86 1.60
48 13 66 26
68.0 11.8 90.0 53.0
1.65 0.05 1.73 1.55
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
309 Tibial longitudinal
Tibial longitudinal axis
D
MTiTh
MDV
C MDA
V
Tibial longitudinal axis
G
A
APG
APA
B SRTi
TiW
H K
J
TiAL
MalW TaAL
E
F
L
SRTa
TaW
M
(b)
(a)
Fig. 1. Sketch of the sagittal (a) and frontal (b) profiles of the tibio-fibular and talar segments. The relevant measurements taken in this work are schematically illustrated (reprinted from Stagni et al. (2004), with permission from Elsevier).
• Trochlea tali length (TaAL): length of the segment connecting the most posterior (F) and the most anterior (E) point of the trochlea tali sagittal arc. • Sagittal radius of the trochlea tali arc (SRTa): radius of the circumference by least square fitting the points of the trochlea tali arc profile.
Test was used, respectively, when relevant variance was assessed to be statistically equal or not. The Pearson correlation coefficient r was also calculated between each measured value and the subjectsÕ height, and for any possible pair of measurements.
The measurements taken from the frontal projection (Fig. 1(b)) were:
3. Results
• Tibial width (TiW): distance of the two intersections (G and H) of the two lines by least square fitting the internal profiles of the two malleoli and the line fitting the top of the tibial mortise. • Malleolar width (MalW): distance along the mediolateral axis between the most medial point of the medial profile (K) and the most lateral point of lateral profile (J) of the tibia/fibula. • Tarsal width (TaW): distance of the two intersections (L and M) of the two lines by least square fitting the medial and lateral talar articular profiles and the line fitting the top of talar articular profile. Mean value, standard deviation, maximum, minimum and median values of all these measurements were calculated over the whole group and also within the two gender groups. The measurements of the two gender groups were compared in order to point out whether they were different, in particular whether those in the male group were larger with respect to the female one. For the comparison, because the normality of the two groups could not be rejected, the T-test or Aspin–Welch
A large range of values for each of these measures was obtained from the subject population analysed. Mean value, standard deviation, maximum, minimum and median values over all the analysed subjects are reported in Table 2, also separated according to gender. Most measurements were larger in the male group than in the female according to the P-values reported in Table 2, except APA, APG, MDV, and MDA. Correlation coefficient r was found to be lower than 0.5 between height of the subject and any of the measurements. The highest correlation between measurements was found for TaAL and SRTa with r = 0.79, and a slightly lower for SRTi and SRTa, r = 0.75. Finally, r = 0.68 was found between TiAL and SRTi, r = 0.62 between TiW and TaW and between TaAL and TaW, and r = 0.58 between TiAL and TiW for the whole group of analysed subjects. Whereas the former are expected because of the same articulating silhouettes, the latter two indicate a nearly constant ratio between width and length respectively at trochlea tali and tibial mortise. No significant correlation was also found between the only one possible external measurement (MalW) and any of the other measurements.
310
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
Table 2 Mean, standard deviation, maximum, minimum, median values for all 12 measurements of the ankle joint, all measurements in mm but APA (deg). Data for the whole group, and for the male and female groups, are reported in the three sections. In the last column on the right it is reported the P-value corresponding to the hypothesis that the measurement in the male group is larger than in the female group All (n = 36)
TiAL SRTi APG APA TiW MalW TaAL SRTa TaW MTiTh MDV MDA
Male (n = 23)
Female (n = 13)
Male > Female
Mean
SD
Max
Min
Median
Mean
SD
Max
Min
Median
Mean
SD
Max
Min
Median
P value (.05)
31.4 27.8 2.7 5.0 31.9 69.0 41.7 23.4 30.4 41.4 8.7 11.5
3.5 4.4 1.8 3.4 3.5 7.6 4.4 3.1 3.3 3.9 3.5 3.5
39.8 34.9 6.8 12.4 40.4 79.6 49.6 32.5 40.2 48.9 18.0 21.4
24.3 21.2 0.0 0.0 25.9 54.0 34.2 19.1 24.2 33.7 3.2 6.7
31.5 26.8 2.5 5.1 32.1 69.9 41.5 23.3 30.1 41.2 8.0 10.7
33.1 29.3 2.6 4.7 33.6 71.0 43.6 24.5 31.5 42.2 9.3 12.2
2.7 4.2 1.6 2.9 2.8 7.4 3.9 3.0 3.5 3.6 2.9 3.2
39.8 41.5 5.5 10.1 40.4 79.6 49.6 32.5 40.2 48.9 16.5 20.4
28.8 23.1 0.1 0.2 27.0 54.0 35.1 19.9 24.2 33.7 5.6 8.0
33.3 27.9 2.6 4.8 33.6 71.9 44.2 24.5 31.9 43.2 8.1 11.2
28.1 24.7 2.7 5.5 28.6 63.5 37.9 21.1 28.3 38.3 7.7 10.4
2.5 4.4 2.3 4.3 2.1 5.1 2.4 1.9 1.4 2.2 4.2 3.9
34.6 32.4 6.8 12.4 31.9 70.3 41.6 26.4 31.0 41.2 18.0 21.4
24.3 21.2 0.0 0.0 25.9 55.1 34.2 19.1 26.2 33.7 3.2 6.7
27.8 24.1 2.4 5.4 29.1 63.4 37.6 20.9 28.1 38.7 7.3 9.6
0.99 0.97 0.01 0.03 0.99 0.91 0.99 0.98 0.97 0.99 0.29 0.14
4. Discussion and conclusions A reliable statistical analysis of ankle joint morphometric measurements obtained from a large population of normal subjects is very important in the design of ankle arthroplasty, particular in sizing prosthesis components and surgical instruments. In the present study, this analysis was performed for the first time on measurements obtained using a previously validated semiautomated tool explicitly designed for the purpose (Stagni et al., 2004). The few present measures that correspond with quantities previously reported in the literature compare quite
well. Present TiAL and TaAL compare well with the sagittal length of the tibial arc (termed ÔCDÕ, mean value 30.8 mm ± 3.6 mm) and the sagittal length of the trochlea tali arc (termed ÔcdÕ, 38.5 mm ± 2.2 mm), previously reported by Fessy et al. (1997). Mean values reported by Mariani and Patella (1977) for tibial and talar width (termed ÔbÕ and ÔdÕ), tibial and talar arc length (ÔgÕ and ÔiÕ) were respectively 35.5 mm, 32.9 mm, 29.9 mm, 36.1 mm. These mean measurements are slightly smaller than those here reported, likely accounted for a larger percentage of female in that study. The consistent larger radius of curvature in the sagittal plane for the tibial than for the talus articulating arcs supports further the evi-
Fig. 2. Diagram showing values of the main dimensions of the current total ankle replacement designs (five points star = STAR, Waldemar Link, Hamburg, Germany; four points star = Buechel-Pappas Total Ankle Prosthesis, Endotec Australia Pty. Ltd., Rosanna, Victoria Australia; diamond = ALBATROS, ProMed—Lepine, Lyon Cedex, France; circle = TNK, Kyocera Corporation, Kyoto, Japan; square = BOX, Finsbury Instruments, Leatherhead Surrey, UK) superimposed to corresponding measurement from the present study (in the form of min–max bar plus the median point). The sizes made available by five replacement designs are reported along the horizontal axis, for the tibial (Ti) and talar (Ta) components, and for the antero-posterior (AP), medio-lateral (ML) dimensions. The acronyms here utilised for the corresponding measure on the intact joint are reported in the diagram in correspondence of the bars. The measurements are all in millimetres.
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
dence of a not full conformity at the talo-crural joint (Leardini, 2001). The present measurements were also compared with available data of corresponding dimensions in the different sizes of the current total ankle replacement designs (Fig. 2). It is interesting to observe that all these designs seem to cover a very limited range and to be in general underestimations of the real ankle dimensions. This may be critical to the full coverage of the cortical bone at the bone cuts by the relevant prosthetic components, necessary to limit the risk of component sinking. The interesting information on posterior narrowing of the tibial mortise cannot be worked out by the technique utilised. As for the surgical technique, the knowledge of the range of values for MDV, MDA, APG, and APA certainly facilitates the design of the optimal level and inclination of the bone saw cuts and of relevant instrumentation in ankle arthroplasty, particularly critical for the necessity of minimising the bone stock to be removed. With dimensions of the sagittal arc of the talus (TaAL and SRTa) a more rational planning of the talar chamfers can be planned. Finally, width of the ankle mortise (TiW) is also important for reduction of the risk of malleolar fracture during and after ankle replacement. With respect to previous morphometric analysis of the ankle joint (Mariani and Patella, 1977; Fessy et al., 1997), the present study reports a larger set of measures, for which the radiologic magnification was also taken into account. These data were collected with a new semi-automated technique, whose reliability and little dependency on the operatorÕs skill were previously assessed experimentally (Stagni et al., 2004). In conclusion, ankle morphometric data reported in the present work define a sizing basis particularly useful for the design of ankle prostheses, and can be useful in the analysis of geometric relations between the anatomical structures of this joint.
311
References Brage, M.E., Bennett, C.R., Whitehurst, J.B., Getty, P.J., Toledano, A., 1997. Observer reliability in ankle radiographic measurements. Foot Ankle Int. 18, 324–329. Coughlin, M.J., Saltzman, C.L., Nunley, J.A., 2002. Angular measurements in the evaluation of hallux valgus deformities: a report of the ad hoc committee of the American Orthopaedic Foot & Ankle Society on angular measurements. Foot Ankle Int. 23, 68–74. Elias, S.G., Freeman, M.A., Gokcay, E.I., 1990. A correlative study of the geometry and anatomy of the distal femur. Clin. Orthop. 260, 98–103. Erkman, M.J., Walker, P.S., 1974. A study of knee geometry applied to the design of condylar prostheses. Biomed. Eng. 9, 14–17. Fessy, M.H., Carret, J.P., Bejui, J., 1997. Morphometry of the talocrural joint. Surg. Radiol. Anat. 19, 299–302. Kempson, G.E., Freeman, M.A., Tuke, M.A., 1975. Engineering considerations in the design of an ankle joint. Biomed. Eng. 10, 166–171. Leardini, A., 2001. Geometry and mechanics of the human ankle complex and ankle prosthesis design. Clin. Biomech. 16, 706–709. Mariani, G., Patella, V., 1977. Valutazione statistica dei parametri articolari dellÕarticolazione tibio-talare. Chir. Organi Mov. 63, 333–340. McCarthy, J.C., Bono, J.V., OÕDonnell, P.J., 1997. Custom and modular components in primary total hip replacement. Clin. Orthop. 344, 162–171. Mensch, J.S., Amstutz, H.C., 1975. Knee morphology as a guide to knee replacement. Clin. Orthop. 112, 231–241. Nakai, T., Takakura, Y., Tanaka, Y., Sugimoto, K., Tamai, S., Kurumatani, N., 2000. Morphologic changes of the ankle in children as assessed by radiography and arthrography. J. Orthop. Sci. 5, 134–138. Prakash, U., Wigderowitz, C.A., McGurty, D.W., Rowley, D.I., 2001. Computerised measurement of tibiofemoral alignment. J. Bone Joint Surg. Br. 83, 819–824. Roberts, S.N., Foley, A.P., Swallow, H.M., Wallace, W.A., Coughlan, D.P., 1991. The geometry of the humeral head and the design of prostheses. J. Bone Joint Surg. Br. 73, 647–650. Stagni, R., Leardini, A., Catani, F., Cappello, A., 2004. A new semiautomated measurement technique based on X-ray pictures for ankle morphometry. J. Biomech. 37, 1113–1118. Tanaka, Y., Takakura, Y., Sugimoto, K., Kumai, T., Sakamoto, T., Kadono, K., 2000. Precise anatomic configuration changes in the first ray of the hallux valgus foot. Foot Ankle Int. 21, 651–656.
Ankle morphometry evaluated using a new semi-automated technique based on X-ray pictures Rita Stagni a
a,*
, Alberto Leardini b, Andrea Ensini b, Angelo Cappello
a
Dipartimento di Electronica, Informatica e Sistemistica, Universita` degli Studi di Bologna, Viale Risorgimento 2, 40136 Bologna, Italy b Laboratorio Analisi del Movimento, Istituti Ortopedici Rizzoli, Via di Barbiano 1/10, 40136 Bologna, Italy Received 1 October 2004; accepted 15 November 2004
Abstract Background. The study of bone morphometry is important in anatomy, biomechanics and the design of joint arthroplasty, because the design of prostheses depends also on the knowledge of joint geometry, both for the replication of function and for the proper sizing of the components. Relevant information for the human ankle joint bones is particularly scarce. Methods. A large number of morphological measurements of the ankle joint were collected in 36 normal subjects by means of a new semi-automated radiographic measurement method based on standard frontal and sagittal X-ray projections. Lengths, heights and widths of the main parts of this articulation together with the radii of curvature of the talar dome and of the trochlea tali in the sagittal plane were worked out. Statistical analysis was performed on these measurements. Findings. A large spectrum for the values of these measures was found. Significant correlation was found among some of the measures, but not between any of these and the subject height. Interpretation. Bones at the ankle joint have similar shapes, i.e. the three dimensions are in direct proportion to each other. No one of the internal dimensions is correlated with the only one possible external measurement, i.e. the malleolar width. Sizes of the current total ankle replacement designs may differ considerably from real corresponding joint dimensions. 2004 Elsevier Ltd. All rights reserved. Keywords: Morphometry; X-rays; Talo-crural joint; Curvature radii; Prosthesis size; Automated measurements
1. Introduction Clinical success of all kinds of human joint arthroplasty depends also on available information on the morphology of the relevant bones. For example, the disappointing clinical results of current ankle implants have been primarily related to unsatisfactory prosthesis design. The design of this replacement relies on the understanding of natural mobility and stability (Leardini, 2001), and thus, on a thorough knowledge of joint geometry. Moreover, in designing human joint prostheses, also the geometry of the components needs atten*
Corresponding author. E-mail address: [email protected] (R. Stagni).
0268-0033/$ - see front matter 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.clinbiomech.2004.11.009
tion, particularly with regards to the different size dimensions. Although several studies for other human joints are reported in the literature (Erkman and Walker, 1974; Roberts et al., 1991; McCarthy et al., 1997), studies about the morphology of the bones forming the ankle or talo-crural joint are scarce. Only a few studies have reported on ankle joint geometrical measurements (Kempson et al., 1975; Mariani and Patella, 1977; Fessy et al., 1997). Although large samples have been analysed by these authors (41, 100 and 50 subjects, respectively), the reliability of the measurements has never been reported, and even magnification not taken into account (Fessy et al., 1997). Accuracy of the method adopted to perform the morphological analysis is a fundamental issue. The method
308
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
sphere (20 mm diameter, 0.01 mm tolerance) in front of the ankle in the lateral projection, and over the lateral malleolus in the antero-posterior projection. This was necessary to calculate X-ray magnification in each patient exactly at the level of the plane under analysis (Stagni et al., 2004). The main general characteristics of these subjects are reported in Table 1. The X-ray pictures were taken in un-loaded conditions when the subject was lying supine on the radiological bed, the foot was in neutral position with respect to the shank and approximately half of the shank and of the foot were visible. Morphological measurements were taken for each pair of X-rays using a software package designed for the purpose. Relevant accuracy was assessed to be, in a previous specifically addressed study (Stagni et al., 2004), within 1 mm and repeatability was not affected by operatorÕs skill. This software allows a series of measurements to be taken on the bone silhouettes, schematically shown in Fig. 1. The measurements taken from the sagittal projection (Fig. 1(a)) were
must allow large groups of subjects to be measured, with minimal invasiveness in case of in vivo studies, and reliable data to be produced in order to perform valuable statistical analysis. Although standard radiograms are routinely used to perform morphological measurements (Mensch and Amstutz, 1975; Mariani and Patella, 1977; Elias et al., 1990; Fessy et al., 1997), the relevant reliability was found to be significantly dependent on the operatorÕs skill (Brage et al., 1997). Although recent studies (Nakai et al., 2000; Tanaka et al., 2000) still measured joint morphometry from radiograms directly by using protractors or goniometers, current recommendations in this respect (Coughlin et al., 2002) pointed out the necessity of awareness. Others (Prakash et al., 2001) already demonstrated the important improvement in terms of accuracy and repeatability of the measurements performed using automatic procedures. A new semi-automatic method based on standard radiograms (Stagni et al., 2004) has been recently proposed and validated for the measurement of a large number of morphometric variables at the human ankle joint. This method was assessed to be accurate, repeatable and little dependant on operatorÕs skill. A relevant statistical analysis of these variables is of fundamental importance for the design and sizing of ankle prostheses and of the relevant surgical instruments. The purpose of this study was to report on a set of morphometrical measurements at the human ankle joint obtained from a large sample of normal subjects by using the novel method (Stagni et al., 2004).
• Tibial arc length (TiAL): length of the segment connecting the most posterior (B) and the most anterior (A) points of the tibial mortise. • Sagittal radius of the tibial mortise (SRTi): radius of the circumference fitting the points of the tibial mortise profile. • Antero-posterior gap (APG): distance along the tibial longitudinal axis between A and B. • Antero-posterior inclination angle of the tibial mortise (APA): angle between the segment AB and the tibial antero-posterior axis. • Maximal tibial thickness (MTiTh): distance between the most anterior point of the tibial anterior profile (C) and the corresponding point (D) along the tibial antero-posterior axis on the posterior tibial profile. • Distance of level of MTiTh from the anterior limit of the mortise (MDA): distance along the tibial longitudinal axis between A and C. • Distance of level of MTiTh from the vertex of the mortise (MDV): distance along the tibial longitudinal axis between the vertex of the tibial mortise V, defined as the intersection between the tibial longitudinal axis and the tibial mortise profile, and C.
2. Methods Lateral and frontal X-rays pictures of the ankle joint were acquired in 36 subjects (13 females, 23 males). These patients were selected among those reporting with an ankle sprain or trauma at the Emergency Room of the authorÕs Institute in the period November 2000–January 2001. Among these, when standard diagnostic imaging procedure was ordered, only those to which bony, joint and syndesmotic integrity was assessed before and after the standard imaging procedure were included in the present analysis. These consented to modify the standard procedure by simply sticking a steel
Table 1 Mean value, standard deviation, maximum and minimum values of age (years), weight (kg) and height (m) of the group of 23 men and 13 women All (n = 36)
Mean SD Max Min
Male (n = 23)
Female (n = 13)
Age (years)
Weight (kg)
Height (m)
Age (years)
Weight (kg)
Height (m)
Age (years)
Weight (kg)
Height (m)
38 17 76 15
71.9 11.1 91.0 52.0
1.72 0.09 1.86 1.55
33 17 76 15
74.0 10.4 91.0 52.0
1.76 0.08 1.86 1.60
48 13 66 26
68.0 11.8 90.0 53.0
1.65 0.05 1.73 1.55
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
309 Tibial longitudinal
Tibial longitudinal axis
D
MTiTh
MDV
C MDA
V
Tibial longitudinal axis
G
A
APG
APA
B SRTi
TiW
H K
J
TiAL
MalW TaAL
E
F
L
SRTa
TaW
M
(b)
(a)
Fig. 1. Sketch of the sagittal (a) and frontal (b) profiles of the tibio-fibular and talar segments. The relevant measurements taken in this work are schematically illustrated (reprinted from Stagni et al. (2004), with permission from Elsevier).
• Trochlea tali length (TaAL): length of the segment connecting the most posterior (F) and the most anterior (E) point of the trochlea tali sagittal arc. • Sagittal radius of the trochlea tali arc (SRTa): radius of the circumference by least square fitting the points of the trochlea tali arc profile.
Test was used, respectively, when relevant variance was assessed to be statistically equal or not. The Pearson correlation coefficient r was also calculated between each measured value and the subjectsÕ height, and for any possible pair of measurements.
The measurements taken from the frontal projection (Fig. 1(b)) were:
3. Results
• Tibial width (TiW): distance of the two intersections (G and H) of the two lines by least square fitting the internal profiles of the two malleoli and the line fitting the top of the tibial mortise. • Malleolar width (MalW): distance along the mediolateral axis between the most medial point of the medial profile (K) and the most lateral point of lateral profile (J) of the tibia/fibula. • Tarsal width (TaW): distance of the two intersections (L and M) of the two lines by least square fitting the medial and lateral talar articular profiles and the line fitting the top of talar articular profile. Mean value, standard deviation, maximum, minimum and median values of all these measurements were calculated over the whole group and also within the two gender groups. The measurements of the two gender groups were compared in order to point out whether they were different, in particular whether those in the male group were larger with respect to the female one. For the comparison, because the normality of the two groups could not be rejected, the T-test or Aspin–Welch
A large range of values for each of these measures was obtained from the subject population analysed. Mean value, standard deviation, maximum, minimum and median values over all the analysed subjects are reported in Table 2, also separated according to gender. Most measurements were larger in the male group than in the female according to the P-values reported in Table 2, except APA, APG, MDV, and MDA. Correlation coefficient r was found to be lower than 0.5 between height of the subject and any of the measurements. The highest correlation between measurements was found for TaAL and SRTa with r = 0.79, and a slightly lower for SRTi and SRTa, r = 0.75. Finally, r = 0.68 was found between TiAL and SRTi, r = 0.62 between TiW and TaW and between TaAL and TaW, and r = 0.58 between TiAL and TiW for the whole group of analysed subjects. Whereas the former are expected because of the same articulating silhouettes, the latter two indicate a nearly constant ratio between width and length respectively at trochlea tali and tibial mortise. No significant correlation was also found between the only one possible external measurement (MalW) and any of the other measurements.
310
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
Table 2 Mean, standard deviation, maximum, minimum, median values for all 12 measurements of the ankle joint, all measurements in mm but APA (deg). Data for the whole group, and for the male and female groups, are reported in the three sections. In the last column on the right it is reported the P-value corresponding to the hypothesis that the measurement in the male group is larger than in the female group All (n = 36)
TiAL SRTi APG APA TiW MalW TaAL SRTa TaW MTiTh MDV MDA
Male (n = 23)
Female (n = 13)
Male > Female
Mean
SD
Max
Min
Median
Mean
SD
Max
Min
Median
Mean
SD
Max
Min
Median
P value (.05)
31.4 27.8 2.7 5.0 31.9 69.0 41.7 23.4 30.4 41.4 8.7 11.5
3.5 4.4 1.8 3.4 3.5 7.6 4.4 3.1 3.3 3.9 3.5 3.5
39.8 34.9 6.8 12.4 40.4 79.6 49.6 32.5 40.2 48.9 18.0 21.4
24.3 21.2 0.0 0.0 25.9 54.0 34.2 19.1 24.2 33.7 3.2 6.7
31.5 26.8 2.5 5.1 32.1 69.9 41.5 23.3 30.1 41.2 8.0 10.7
33.1 29.3 2.6 4.7 33.6 71.0 43.6 24.5 31.5 42.2 9.3 12.2
2.7 4.2 1.6 2.9 2.8 7.4 3.9 3.0 3.5 3.6 2.9 3.2
39.8 41.5 5.5 10.1 40.4 79.6 49.6 32.5 40.2 48.9 16.5 20.4
28.8 23.1 0.1 0.2 27.0 54.0 35.1 19.9 24.2 33.7 5.6 8.0
33.3 27.9 2.6 4.8 33.6 71.9 44.2 24.5 31.9 43.2 8.1 11.2
28.1 24.7 2.7 5.5 28.6 63.5 37.9 21.1 28.3 38.3 7.7 10.4
2.5 4.4 2.3 4.3 2.1 5.1 2.4 1.9 1.4 2.2 4.2 3.9
34.6 32.4 6.8 12.4 31.9 70.3 41.6 26.4 31.0 41.2 18.0 21.4
24.3 21.2 0.0 0.0 25.9 55.1 34.2 19.1 26.2 33.7 3.2 6.7
27.8 24.1 2.4 5.4 29.1 63.4 37.6 20.9 28.1 38.7 7.3 9.6
0.99 0.97 0.01 0.03 0.99 0.91 0.99 0.98 0.97 0.99 0.29 0.14
4. Discussion and conclusions A reliable statistical analysis of ankle joint morphometric measurements obtained from a large population of normal subjects is very important in the design of ankle arthroplasty, particular in sizing prosthesis components and surgical instruments. In the present study, this analysis was performed for the first time on measurements obtained using a previously validated semiautomated tool explicitly designed for the purpose (Stagni et al., 2004). The few present measures that correspond with quantities previously reported in the literature compare quite
well. Present TiAL and TaAL compare well with the sagittal length of the tibial arc (termed ÔCDÕ, mean value 30.8 mm ± 3.6 mm) and the sagittal length of the trochlea tali arc (termed ÔcdÕ, 38.5 mm ± 2.2 mm), previously reported by Fessy et al. (1997). Mean values reported by Mariani and Patella (1977) for tibial and talar width (termed ÔbÕ and ÔdÕ), tibial and talar arc length (ÔgÕ and ÔiÕ) were respectively 35.5 mm, 32.9 mm, 29.9 mm, 36.1 mm. These mean measurements are slightly smaller than those here reported, likely accounted for a larger percentage of female in that study. The consistent larger radius of curvature in the sagittal plane for the tibial than for the talus articulating arcs supports further the evi-
Fig. 2. Diagram showing values of the main dimensions of the current total ankle replacement designs (five points star = STAR, Waldemar Link, Hamburg, Germany; four points star = Buechel-Pappas Total Ankle Prosthesis, Endotec Australia Pty. Ltd., Rosanna, Victoria Australia; diamond = ALBATROS, ProMed—Lepine, Lyon Cedex, France; circle = TNK, Kyocera Corporation, Kyoto, Japan; square = BOX, Finsbury Instruments, Leatherhead Surrey, UK) superimposed to corresponding measurement from the present study (in the form of min–max bar plus the median point). The sizes made available by five replacement designs are reported along the horizontal axis, for the tibial (Ti) and talar (Ta) components, and for the antero-posterior (AP), medio-lateral (ML) dimensions. The acronyms here utilised for the corresponding measure on the intact joint are reported in the diagram in correspondence of the bars. The measurements are all in millimetres.
R. Stagni et al. / Clinical Biomechanics 20 (2005) 307–311
dence of a not full conformity at the talo-crural joint (Leardini, 2001). The present measurements were also compared with available data of corresponding dimensions in the different sizes of the current total ankle replacement designs (Fig. 2). It is interesting to observe that all these designs seem to cover a very limited range and to be in general underestimations of the real ankle dimensions. This may be critical to the full coverage of the cortical bone at the bone cuts by the relevant prosthetic components, necessary to limit the risk of component sinking. The interesting information on posterior narrowing of the tibial mortise cannot be worked out by the technique utilised. As for the surgical technique, the knowledge of the range of values for MDV, MDA, APG, and APA certainly facilitates the design of the optimal level and inclination of the bone saw cuts and of relevant instrumentation in ankle arthroplasty, particularly critical for the necessity of minimising the bone stock to be removed. With dimensions of the sagittal arc of the talus (TaAL and SRTa) a more rational planning of the talar chamfers can be planned. Finally, width of the ankle mortise (TiW) is also important for reduction of the risk of malleolar fracture during and after ankle replacement. With respect to previous morphometric analysis of the ankle joint (Mariani and Patella, 1977; Fessy et al., 1997), the present study reports a larger set of measures, for which the radiologic magnification was also taken into account. These data were collected with a new semi-automated technique, whose reliability and little dependency on the operatorÕs skill were previously assessed experimentally (Stagni et al., 2004). In conclusion, ankle morphometric data reported in the present work define a sizing basis particularly useful for the design of ankle prostheses, and can be useful in the analysis of geometric relations between the anatomical structures of this joint.
311
References Brage, M.E., Bennett, C.R., Whitehurst, J.B., Getty, P.J., Toledano, A., 1997. Observer reliability in ankle radiographic measurements. Foot Ankle Int. 18, 324–329. Coughlin, M.J., Saltzman, C.L., Nunley, J.A., 2002. Angular measurements in the evaluation of hallux valgus deformities: a report of the ad hoc committee of the American Orthopaedic Foot & Ankle Society on angular measurements. Foot Ankle Int. 23, 68–74. Elias, S.G., Freeman, M.A., Gokcay, E.I., 1990. A correlative study of the geometry and anatomy of the distal femur. Clin. Orthop. 260, 98–103. Erkman, M.J., Walker, P.S., 1974. A study of knee geometry applied to the design of condylar prostheses. Biomed. Eng. 9, 14–17. Fessy, M.H., Carret, J.P., Bejui, J., 1997. Morphometry of the talocrural joint. Surg. Radiol. Anat. 19, 299–302. Kempson, G.E., Freeman, M.A., Tuke, M.A., 1975. Engineering considerations in the design of an ankle joint. Biomed. Eng. 10, 166–171. Leardini, A., 2001. Geometry and mechanics of the human ankle complex and ankle prosthesis design. Clin. Biomech. 16, 706–709. Mariani, G., Patella, V., 1977. Valutazione statistica dei parametri articolari dellÕarticolazione tibio-talare. Chir. Organi Mov. 63, 333–340. McCarthy, J.C., Bono, J.V., OÕDonnell, P.J., 1997. Custom and modular components in primary total hip replacement. Clin. Orthop. 344, 162–171. Mensch, J.S., Amstutz, H.C., 1975. Knee morphology as a guide to knee replacement. Clin. Orthop. 112, 231–241. Nakai, T., Takakura, Y., Tanaka, Y., Sugimoto, K., Tamai, S., Kurumatani, N., 2000. Morphologic changes of the ankle in children as assessed by radiography and arthrography. J. Orthop. Sci. 5, 134–138. Prakash, U., Wigderowitz, C.A., McGurty, D.W., Rowley, D.I., 2001. Computerised measurement of tibiofemoral alignment. J. Bone Joint Surg. Br. 83, 819–824. Roberts, S.N., Foley, A.P., Swallow, H.M., Wallace, W.A., Coughlan, D.P., 1991. The geometry of the humeral head and the design of prostheses. J. Bone Joint Surg. Br. 73, 647–650. Stagni, R., Leardini, A., Catani, F., Cappello, A., 2004. A new semiautomated measurement technique based on X-ray pictures for ankle morphometry. J. Biomech. 37, 1113–1118. Tanaka, Y., Takakura, Y., Sugimoto, K., Kumai, T., Sakamoto, T., Kadono, K., 2000. Precise anatomic configuration changes in the first ray of the hallux valgus foot. Foot Ankle Int. 21, 651–656.