On the structural symmetry of human femurs

Bone, 8, 165-169 (1987) Printed in the USA. All rights reserved

Copyright

8756-3282187 $3.00 + .OO 0 1987 Pergamon Journals Ltd.

On the Structural Symmetry of Human Femurs M. PORTIGLIATTI-BARBOS,l

S. CARANDO,’

A. ASCENZl,2 and A. BOYDE3

1 Department of Orthopaedics, University of Torino, Via Zurett/, 29, l-10726, Tonno, Italy. 2 Department of Human Bio-pathology, Section of Morbid Anatomy, “La Sapienza” Uruverslty, Vlale Regina Elena 324, Polic/!n!co, l-001 61, Rome, Italy. 3 Department of Anatomy and Embryology, Unlverssity College London, Gower Street, London WC1 E 6BT, England Address for correspondence I-10126, Tonno, Italy

and reprints: M. Portigliatti-Barbos, Department of Orthopaedics, University of Torino, Via Zuretti, 29,

cated that bone structures in femoral shafts are symmetrically distributed on the right and left sides, even if a certain degree of variation was found at the level of osteons (second-order structures) and lamellae (third-order structures). Later, another important feature was demonstrated by Marotti (1961, 1963), and Amprino and Marotti (1964). Remodelling in compact bone is virtually identrcal at the same levels in the right and left long bones. These results were corroborated by those of Barasa and Shahar (1966), who were able to show that in femurs and tibiae of dogs a close symmetry exists at the microscoprc level between left and right bones, and that such symmetry becomes more and more marked as the animals age. Finally, fluff and Hayes (1963) have described sex, age and size differences in the cross-sectional geometry of Pecos Pueblo femora and tibiae. The aim of the present investigation was to determine whether the two femora from one indivtdual would show symmetrically distributed bone structures and whether they would show a symmetrical rotatronal distribution of the two types of lamellae (transverse and longitudinal) along the shaft. The bone material for the present study was obtained from the skeleton of a woman, and this study can be considered an extension of our previous work (Portigliatti-Barbos et al., 1984) on the structure and blomechanics of the male femur

Abstract The distribution of transverse and longitudinal lamellae from osteons and interstitial bone of the two femurs from a woman aged between 25 and 30 was investigated. Exactly plane parallel cross-sections (100 km) were prepared from the upper, middle and lower thirds of the two bones. The distribution of transverse and longitudinal lamellae was determined for both osteonic and interstitial bone, using circularly polarised light. The results show that (a) as in the male femur, the transverse and longitudinal lamellae in these female femurs have a characteristic rotational distribution along the shaft consistent with the distribution of the bending strains normally operative in bone; and (b) the rotational distribution of the two types of lamellae is slightly asymmetrical in the two female bones, probably because of imperfectly symmetrical use of the lower limbs.

Key Words: Femoral Bone Remodelling-Image Disease.

Shaft--Osteon-Bone-LamellaAnalysis-Biomechanrcs-Bone

Introduction

Materials and Methods

Ascenzi and coworkers (Ascenzi et al., 1966, 1985; Ascenzi and Bonucci, 1967, 1968, 1976) have stated that, in accordance with Gebhardt’s theory, secondary osteons and interstitial bone consist of two types of lamellae: longrtudinal lamellae which withstand loading by tension, and tranverse lamellae which withstand loading by compression, Recently Portigliatti-Barbos et al. (1983, 1984) and Boyde et al. (1984), have investigated the distribution of “transverse” lamellae (collagen and crystallites lie parallel to the plane of the transverse section) and “longitudinal” lamellae (collagen and crystallites lie perpendicular to the plane of the transverse section) in human femoral shafts. In this way, the authors provided evidence that the transverse and longitudinal lamellae from osteonic and rnterstrtial bone in the femoral shaft show a characteristic rotational distribution consistent with the distribution of the predicted bending forces operative in this bone. In addition, earlier investigations (Amprino and Sisto, 1946) indi-

The two femurs were obtarned from the skeleton of a woman of unknown age, and were donated by Professor F Mallegni of the Department of Archaeological Sciences, Section of Human Palaeontology, at the University of Pisa. It was established by anthropological methods that the donor was 25 to 30 years old The two femurs were not exactly symmetrical (Fig. 1).The left femur was somewhat longer than the right Transverse slices (1 cm) were cut from each femur by sawing. They corresponded to the mid-points of the upper, middle and lower thirds of the shaft From each slice a serves of complete cross sections (100 pm) was prepared using a Lertz annular blade mrcrotome capable of cutting very plane parallel specimens This is an important point, because under the polarising microscope the difference in the optical path between the extraordinary and ordinary ray is a function of the thickness of the section (Boyde et al., 1984). In each series of sections the one with the most uniform thrckness was used For this reason the corresponding sectrons in the two bones cannot be considered as comrng from exactly the same level 165

166

M. Portigliatti-Barbos

et al

: Structural and biochemical

1. The two femurs studied. The levels at which the sections were prepared

Soft tissue debris was removed from the sections by treatment with a 1% solution of Tergazyme in distilled water at 40°C with occasional ultrasonication. The sections were then mounted in DPX. The linea aspera and outer limits of each section were used to make the orientation of the left and right sections as similar as possible. The proportion of the field occupied by transverse lamellae (bright under all rotations of the section when using circularly polarised light in the polarising microscope) was determined using a Quantimet 720 image-analysing computer with a 2-D detector To correct for the amount of bone tissue present in each field, the area appearing bright under darkfield illumination was measured. The numerical results were processed to yield a map showing the distribution of lamellae in each section To do this, the mean and standard deviation for all bone fields in a section was first determined, after which values for single fields (or groups of fields) were plotted using symbols graded according to the number of half standard deviations on either side of the mean (Boyde et al., 1984).

symmetry

of femurs

are ldic :ated

Results The diagrams on the left side (ace) were obtained from the right femur, and those on the right side (b,d,f) from the left femur (Fig. 2). Going from top to bottom, the three pairs of diagrams (ab,cd,ef) are representative of the upper, middle and lower thirds of the two femurs, respectively. As regards orientation, the left and lower sides correspond to the lateral and anterior aspects of the sections, respectively. For the diagrams on the left side (a,c,e), the left and lower sides correspond to the medial and anterior aspects of the sections. In each diagram the proportion of transverse lamellae is indicated by the size of the plotting symbol. The center of each square is at the center of nine square microscopic fields, each measuring 620 pm. Each square is centered in the position occupied in the bone section. Thus a diagrammatic but morphologically accu-

167

M. Portigliattf-Barbos et al.. Structural and biochemical symmetry of femurs

a

b

d

e

n Ree=Q9a

aEla@Je~plea

f

a~mBBa@BM~~rl

R~~~Eu~ere~m

maxl

~~~m~claan~~~

claaEJEl!BIn ElePeae q E3Igl

q pp4 EBB

EclB 8

tie

ii;

RrJ

oD!l

q

mbil

SPls)

BP0

;:

q l3 emu men

m

l

=

8

Ixlm lxls q

UpqR

D

l

o==oooou88B88fl fdrBe~n3~Ba~~ oBsePeE3D

q er3

PII

tam

Rrn

em ataa

PP= ml9

napPa

880

rmmmBmmnm

q 9emsm8Kl 0

Fig. 2. Diagrams a, b, c, d, e, f display the distribution of transverse and longitudinal lamellae in the sections prepared from the two femurs at the levels indicated in Fig. 1. top = posterior, below = anterior, towards center-line of page = medial, towards edge of page = lateral. Scale: distance between centers of adjacent plotting symbols = 1.86 mm. The size of plotting symbol is proportional to ratio of bright area in circularly polarised light to bright area in dark field illumination.

rate representation is given of an entire section with visualization of the overail ratio between the two types of lamellae. Squares without diagonals mean that the bone section in question covered less than five of the possible

nine microscopic fields. Five or more hits per field are shown by crosses in the square symbol. Fields with less than three of the possible nine were excluded (PortigliattiBarbos et al., 1984).

168

M. Portigliatti-Barbos et al.. Structural and brochemical symmetry of femurs

In the section from the upper third of the right femoral shaft the incidence of transverse lamellae is particularly high in the anterior portion of the medial wall at the boundary with the anterior wall of the bone (Fig. 2a). There is also a concentration of transverse lamellae in a small zone lying near the boundary between the lateral and the posterior wall of the bone, and another small zone on the medial portion of the posterior wall; seen as a whole, high concentrations appear to follow an axis running from the postero-anterior to the lateral-medial side. This obliquity in the distribution of transverse lamellae corresponds to the elliptical shape of the section whose major axis has a similar lateral-medial and postero-anterior direction. In general, there is no clear fall in the incidence of transverse lamellae from the inner to the outer circumference. In contrast, Figure 2b shows an almost rounded shape. The highest concentration of transverse lamellae is found in the medial wall of the femoral diaphysis. A localized concentration is almost appreciable in the inner portion of the lateral wall. A very high, but circumscribed concentration of transverse lamellae is present along the inner border of the anterior wall. As a whole, this distribution reveals a lateral-medial orientation. In this section a fall in the incidence of transverse lamellae from the inner toward the outer circumferential edge may be seen. In Figure 2c there is again a high concentration of transverse lamellae in the medial wall of the femoral diaphysis. They do, however, show a slight but clear tendency to backward displacement, since a concentration can be clearly made out along the inner border of the posterior wall. Within the anterior, lateral and posterior walls there is a fall in the incidence of transverse lamellae from the inner to the outer edge of the section. This finding is least noticeable in the medial wall, where the concentration of the transverse lamellae is almost homogenous. In Figure 2d the distribution of transverse lamellae does not differ much from that shown in Figure 2c. Here again they are concentrated in the medial wall of the diaphysis, but there is no clear tendency to backward displacement. A localized high incidence of transverse lamellae is found at the center of the inner edge of the lateral wall. Over the remainder of the section there is nothing of special significance to record. Along the outer edge of the anterior, posterior and lateral walls, the incidence of transverse lamellae falls sharply. The distribution of transverse lamellae shown in Figure 2e reveals that the highest concentrations are found in the posterior portion of the medial wall and along the posterior wall. In these sites, however, the distribution of transverse lamellae is not homogeneous; it is made rather irregular by alternation with foci in which longitudinal lamellae are prevalent. Away from the outer edge of the section there is a sharp fall in the incidence of transverse lamellae. Conversely, Figure 2f shows that transverse lamellae are most homogeneously concentrated in the posterior wall and in the posterior portion of the medial wall at its boundary with the posterior wall. A fall in the concentration of transverse lamellae from the inner toward the outer edge of the section can only be detected at the level of the posterior wall; in the other walls it is apparently absent.

Discussion We begin by comparing the present data for two female femurs with our previous data for a male (Portigliatti-

Barbos et al., 1984). As regards the upper level, transverse lamellae are preferentially distributed in the medial wall in the right femur of the woman and in the femur of the man, but in the former they are found appreciably nearer the anterior wall. In the male femur there was a concentration of transverse lamellae in a small zone near the boundary between the lateral and the anterior wall: in the female femurs, the concentration of transverse lamellae IS localized in two small zones, lying at the boundary between the lateral and the posterior wall and in the medial portion of the posterior wall. Comparing the upper third, it IS clear that the distribution of transverse lamellae IS very similar to the two bones. The highest concentration is found in the medial wall of the diaphysis, while a less marked concentration is appreciable in the inner side of the lateral wall. A hrgh, but circumscribed concentration is found along the inner border of the anterior wall in the male femur. At the middle level of the shaft, the distribution of transverse lamellae is somewhat simrlar in the right femurs of both sexes. In the male, the highest concentrations of transverse lamellae are found in the posterior portion of the medial wall, at the boundary with the posterior wall. This is also true of the right femur of the female, but in this case a high concentration of the same lamellae can also be observed along the inner border of the posterior wall. In comparing the right male femur with the left bone of the female, the transverse lamellae in the latter are again found to be concentrated in the medial wall of the diaphysis, but there is no obvious tendency to backward displacement. A concentration of lamellae is, in fact, found along the inner border of the posterior wall. Coming now to the lower level of the femoral diaphysis, transverse lamellae in the male femur were exclusively concentrated in the transitional zone between the posterior portion of the medial wall and a large portion of the posterior wall, whereas high concentrations are found in the right female femur both In the posterior portion of the medial wall and along the posterior wall, even if some discontinuities appear in these two sites as a result of alternation with foci of mainly longitudinal iamellae. By contrast, comparison between the femur of the man and the left femur of the woman reveals that, in the latter, transverse lamellae are homogeneously concentrated in the posterror wall, and in the posterior portion of the medial wall at the boundary with the posterior wall. Finally, as in the male femur, most of the sections cut from the two female femurs show that the incidence of longitudinal lamellae tends to rise in the outer parts of these bones. A comparison between the figures showing the distribution of transverse lamellae in the one male and the two female femurs shows that, despite some differences, concentration of transverse lamellae swings from the medial toward the posterior wall in moving from the proximal to the distal end of the shaft in all three diaphyses, whereas, as would be expected, the orientation of longitudinal lamellae swings in the opposite direction, from the lateral toward the anterior wall. Such a change suggests a characteristic distribution of bending forces in the femur (PortigliattiBarbos et al., 1984). The other differences in the distribution of the two types of lamellae in the femur from the man, and in the two femurs from the woman may be due to sexual dimorphism. This is also the opinion of Ruff and Hayes (1983), who suggest two possible explanations for the results of their investigations on general sex differences in the structure of the lower limb bones. The first

M Portigliatti-Barbos et al.: Structural and biochemical symmetry of femurs

involves sexual dimorphism in the structure of the pelvis, while they suppose that the second is related to a probable behavioural difference between males and females in their series of Pecos Pueblo bones. Comparison between the diagrams recorded for the male femur and those for the two female femurs has revealed that differences also exist between the two latter bones. Apart from the possibility that some dissimilarities may be due to inaccuracies in preparing sections from what should be exactly corresponding levels because the two sections were not perfectly symmetrical, there may be some variations in the distribution of bone structures of the second and third orders (osteons and lamellae) in the two symmetrical shafts of the same subject. The investigations of Amprino and Sisto (1946) are enlightening on this point. These authors studied the distribution of the microscopic structures in both femurs of four individuals aged 3, 16, 30 and 83. A close resemblance was found between the two sides as regards the topography and extent of fundamental concentric systems, resorption cavities and osteons, whereas the dimensions of osteons and interstitial bone, and the orientation of collagen bundles in lamellae showed dissimilarities between sections taken from precisely the same levels. These results concur with those of Barasa and Shahar (1966), who found strong similarities between sections from corresponding levels of the left and right femurs and tibiae of dogs by studying the number and diameters of osteons, the amount of primary and secondary bone, and the porosity of the material. The results of our investigation yield evidence that, in spite of local asymmetrical arrangements, the cumulative distribution within the various bone structures in terms of cross-sectional areas taken up by transverse lamellae reveals a symmetrical arrangement of the lamellae themselves corresponding to a swing in orientation from the medial toward the posterior wall in moving from the proximal to the distal end of the shaft. It would be interesting to have information about the degree of microscopic asymmetry of two homologous long bones pertaining to the same individual when one of them undergoes a noticeable hypertrophy in response to exercise, as in the group of professional tennis players reported by Jones et al. (1977). The cortical thickness of the humerus on the playing side was greater by 34.9% in men and 28.4% in women compared with the control side. In conclusion, our findings show the same rotational arrangement of bone lamellae along the diaphysis as in the male femur previously investigated, and tend to show that this arrangement can be considered as symmetrically distributed in the two bones of the same individual, even If some differences have arisen from an asymmetrical use of the lower limbs.

169

Achnowledgement: The present study has been supported by grants from the National Research Council of Italy (NO. 83.02541.04 and 84.02275 04) and the Medical Research and the Science Research Council, UK The Authors thank Dr Lidia Salvadorl for helpful technical assistance in preparing bone sections

References Amprlno R and Marotti G A topographic quantitative study of bone formation and reconstructfor- In. Bone and Tooth. H J J Blackwood. ed Pergamon Press, Oxford, 1964 Amprino R and Sisto L Analogies et differences de structure dans les dlfferentes regions d’un meme OS. Acta Anat 2.202-214. 1946 Ascenzi A and Bonucci E The tensile properties of single osteons Anaf Ret 158 375-386. 1967 Ascenzl A and Bonucc E. The compressive propertIes of single osteons. Anat Ret 161 377-392, 1968 Ascenzi A and Bonucci E RelatIonship between ultrastructure and pin test in osteons C/in. Orfhop. 121 275-294, 1976 Ascenzl A Bonuccl E and Checcuccl A The tensile properties of single osteons studled using a microwave extenslmeter In Studies on the Anatomy and Function of Bone and Jo/n& F G Evans, ed Springer, Berlin, pp 121-141, 1966 Ascenzi A. Benvenutl A, Mango F and SlrnllI R MechanIcal hysteresis loops from single osteons TechnIcal devtces and preliminary results J Biomechanu 18:391-398, 1985 Barasa A and Shahar A Symetrie structurale des OS pairs homologues chez les mammlferes &ii. Ass. Anal. 51 123-128, 1966 Boyde A Bianco P Portigliatti-Barbos M and Ascenzl A Collagen orlentatlon cn compact bone I A new method for the determination of the proportion of collagen parallel to the plane of compact bone sections Metab Bone DIS Rel. Res. 5 299-308. 1984 Jones H H Priest J D Hayes WC Tichenor CC and Nagel D.A Humeral hypertrophy In response to exercise J Bone & dornt Surg 59-A 204-208 1977 Marottl G Number and arrangement of osteons In corresponding regions of homotlplc long bones Nature 191 1400-1401, 1961 Marotti G Quantltatlve studies on bone reconstructIon 1 The reconstructlon in homotlpic shaft bones Acta Anaf 52 291-333, 1963 Portlgllattl-Barbos M Blanc0 P and Ascenzl A Dlstrlbutlon of osteonlc and lnterstltlal components In the human femoral shaft with reference to structure, calclflcatlon and mechantcai propertles Acta Anat 115 178-186 1983 Portlgllattl-Barbos M Blanc0 P Ascenzl A and Boyde A Collagen orientatlon in compact bone II Dlstrlbution of lamellae In the whole of the human femoral shaft with reference to its mechanical properties Metab Bone D/s Rei Res 5 309-315, 1984 Ruff C B and Hayes WC Cross-sectIonal geometry of Pecos Pueblo femora and tibiae--A blomechanical investlgatlon II Sex, age, and side differences Amer J Phys Anthropoi 60 383-400. 1983

Recerved. June 19, 7986 Rewsed November 6, 1986 Accepted November 28. 1986