F
`
`my...
`
`_
`
`1305 Linden Dr, Madison. W . 53706
`
`Vol. 16, No. 12. 1983
`
`List of Contents and Author Index for Volume 16
`bound in at the and of this issue
`
`
`AL‘K‘I"_'l‘—Ik
`
`JOURNAL
`I III:
`OF
`'
`BIOMECHANICS
`
`_
`Editors 4:} -Chr'ef
`Verne L. Roberts and Rik Huiskes
`
`Published by
`PEHGAMON PRESS
`
`
`
`Oxiord ‘ New York - Toronto ‘ Sydney - P'aris - Frankfurt
`
`Page 1 0f 15
`
`ZIMMER EXHIBIT 1015
`
`Page 1 of 15
`
`ZIMMER EXHIBIT 1015
`
`

`

`Journal of Biomechanics
`Affiliated with the American Society of Biomechanics and the European Society of Biomechanics
`Editors-in-Chief
`Verne L. Roberts
`institute for Product Safety. 1410 Duke University Road. Durham. North Carolina ZFTDI. USA
`Rik Huiskes
`Biomechanics Laboratory. Department oi Orthopaedics. University of Niimegen, 6500 HS Niimegen. The Netheriands
`Editor Emeritus
`F. Gaynor Evans
`Department of Anatomy, University of Michigan. Ann Arbor. Michigan 48104, USA
`Editorial Board
`Richard A. Brand
`Department of Orthopaedic Surgery. The University of laws. iowa City, M 52242. USA
`Dick H. van Campen
`Department of Mechanical Engineering. Twente University oi Technology. 7500 AE Enschede. The Netherlands
`Victor H. Frankel
`Hospital for Joint Diseases. Orthopaedic institute. 301I East 17th Street. New York, NY 10003. USA
`Werner Goldsmith
`Coiiege of Engineering. Mechanical Engineering. University of Caiifornia. Berkeley. CA 97420. USA
`Stephan M. Perren
`Laboratory for Experimentai Surgery. Schweizerisches Forschungsinstitut. CH~Davos 727'0. Switzeriand
`Albert E. Schultz
`Department oi Mechanicai Engineering and Applied Mechanics. University of Michigan. Ann Arbor. Mi 43103, USA
`Editorial Consultants
`James G. Andrews. University oi iowa. iowa City. USA
`Antonio Ascenzi, Universita di Roma. Horne.
`itar‘y
`Rene Bourgois. Ecoie Royaie Miiitaire. Srusseis. Beigium
`Charles J. Burstone. University of Connecticut. Farm—
`ington. CT, USA
`Aurelio Cappozzo. Universita degii Studi di Roma. Home.
`itaiy
`Donald Chafiin. University oi Michigan. Ann Arbor. Mi.
`USA
`Edmund Y. 5. Chap. Mayo Ciinic, Rochester. MN. USA
`Roy D. Crowninshield. The University oi iowa. iowa City.
`USA
`John D. Currey. University of York. York. United
`Kingdom
`Paul Ducheyne. University oi Leuven. Heyeriee. Beigiur‘n
`Henning E. van Gierke, Wright Patterson Air Force Base.
`OH. USA
`Muzio M. Gola. Poiitecnico di Torino. Torino. itaiy
`James G. Hey. The University at iowa. iowa City. USA
`SA
`Wilson C. Hayes. Harvard Medicai Schooi. Boston. MA.
`J. Lawrence Katz. Rensseiaer Poiytechnic institute. Troy,
`NY, USA
`Albert l. King. Wayne State University. Detroit. Wi. USA
`Reinhard Kblbel. Orthopédische Universitaitsiriiniir. Ham -
`burg. West Germany
`Lance E. Lanyon. Tufts University. North Grafton. MS,
`USA
`James H. McElhaney. Duke University. Durham. NC.
`USA
`Bernard F. Money. Mayo Ciinic. Rochester. MN. USA
`Annual Subscription Rates 1984 {including postage and insurance)
`Published monthly
`For libraries. university departments, government laboratories. industrial and other multiple-reader institutions US $230.00. ‘2 year rate
`$513.00. Specially Reduced Rate to Individuals: in the interacts of maximizing the dissemination of the research results published in
`this important international journal we have established a two-tier price structure. Any individual. whose institution takes out a library
`subscription. may purchase a second or additional subscription for personal use at a much reduced rate of US $55.00. Airmail
`subscription extra. Subscription enquiries from customers in North America should be sent to; Pergamon Press Inc.. Maxwell House.
`OX3 03W, U.
`.
`Fairview Park.KE!msford. NY 10523. USA, and for the remainder of the world to; Pergamon Press Ltd.. Headington Hill Hall, Oxford
`Microiorm Subscriptions and Back issues
`Back issues of all previously published volumes are available in the regular editions and on microfilm and microfiche. Current
`subscriptions are available on microfiche simultaneously with the paper edition and on microfilm on completion of the annual index
`at the and ol' the subscription year.
`Publishing and Advertising Offices: PERGAMON PRESS
`Headington Hill Hall. Dxl'ord OX3 03W, England
`Maxwell House. Fairview Park. Elmslord. New York 10523. USA
`Copyright © 1983 Pergamon Press Ltd.
`it is a condition of publication that manuscripts submitted to this journal have not been published and will not be simultaneously
`submitted or published elsewhere. By submitting a manuscript, the authors agree that the copyright for their article is transferred to
`the publisher it and when the article is accepted for publication. However. assignment of copyright is not required from authors who
`work tor organizations which do not permit Such assignment. The copyright covers the exclusive rights to reproduce and distribute
`the article. including reprints. photographic reproductions. microform or any other reproductions of similar nature and translations. NO
`part of this publication may be reproduced. stored in a retrieval system or
`transmitted in any form or by any means, electronm.
`electrostatic. magnetic tape. mechanical. photocopying. recording or otherwise. without permission in writing from the copyright
`holder.
`
`Vafl‘gé Mow. Rensseiaer Polytechnic institute, Troy. NY,
`R033: J. Nikolai. St. Louis University, St. Louis. MO.
`Abraham Noordergraaf. University of Pennsyhranie, Phii-
`adeiphia, PA. USA
`Y. C. Pao. The University oi Nebraska. Lincoin. USA
`John P. Paul, University oi Strathciyde. Giesgow. United
`Kingdom
`Raymond J. Pearson. The University of Michigan. Ann
`Arbor. Mi, USA
`Malcolm H. Pope. The University oi Vermont. Buriington.
`UT, USA
`Carol Putnam. Daihousie University. Haiiiax, Canada
`Daniel Ouemada. Universite Paris. Paris, France
`John T. Scales. Royai National Orthopaedic Hospital.
`Stanrnore. United Kingdom
`Goran Solvik. University of Land, Lund. Sweden
`Ian Stokes. University of Vermont. Buriington, liT, USA
`Andrus \iiidik. University of Aarhus. Aarhus. Denmark
`Raymond P.
`\iito, Georgia institute oi Technoiogy.
`Atlanta. GA. USA
`Stflagjn A. Wainwright. Duke University, Durham. NC.
`Peter S. Walker. Veterans Administration Medical Center,
`Boston. MA. USA
`Herman J. Woitring. Overasseir. The Netheriands
`Savio L-Y Woo. University oi Caiii'ornia. La Jolie, CA,
`USA
`Donald F. Young, iowa State University. Ames. iA. USA
`
`U.S. Copyright Law applicable to users in the USA
`Photocopying iniorrnation ior users in the USA: The Item-lee Code for this publication Indicates that authorization to photocopy
`items ior internal or personal use is granted by the copyright holder for libraries and other users registered with the Copyright Clearance
`Center {CC-C) Transactional Reporting Service provided the stated tea for copying beyond that permitted by Section 10? or 103 of
`the United States Copyright Law is paid. The appropriate remittance ct $3.00 per copy per article is paid directly to the Capynght
`Clearance Center Inc.. 21 Congress Street. Salem. MA 01970. The copyright owner‘s consent does not extend to copying for genera1
`distribution. for promotion. for creatlng new works, or for resale. Specific written permission must be obtained from the publisher for
`such copying.
`in case of doubt please contact your nearest Pergamon office.
`The item -Fee Code for this pubir'catr'on is: 0021—9290i83 $3.00 + 0.00
`
`Page 2 of 15
`
`Page 2 of 15
`
`

`

`Its. No. I}. pp. WI J’s]. :93}.
`.r. Himft'mlf'.‘ Vol.
`printed in Brain Britain
`
`l‘ll'lll -92‘30-R.\ $1.00 + .llltl
`I‘ll?! Pcryumon Press thl.
`
`.l"
`
`THE MECHANICAL CHARACTERISTICS OF CANCELLOUS
`BONE AT THE UPPER FEMORAL REGION
`
`M. Mann-ms". R. VAN AUDEKERCKET, P. DELPORT‘. P. DE MEESTER‘l‘ and J. C. MULIER"
`‘Acadernisch Ziekenhuis. B 304] Pellenberg, Belgium; flCOBl. Biomechanics and Biomaterial Section.
`K. U. Leuven. Belgium
`
`Abstract—Mechanical behaviour of Lrabecular bone at the upper femoral region of human bones has been
`studied by compression tests on trabecular bone specimens removed from normal femurs obtained at
`autopsy. Compression tests were performed along three different axes of loading on wet specimens and high
`loading rates. Femoral head specimens proved to be the strongest for any axis of loading,
`Large variation in compressive strength and modulus ofelasticily is seen within and between femoral bone
`samples. Anisotropy and differences in anisotropy for the different regions have been observed. A significant
`correlation between mechanical properties (omax -—E) and bone mineral content of the specimen was found.
`Tests on whole bone structures demonstrate that removal of the central part of the trabecular bone at the
`proximal femur reduces the strength for impact loading considerably {: 503;].
`
`lNTRODUCfiON
`
`Cancellous (trabecular. spongy} bone is an open cell
`porous structure which is present at the epiphyseal and
`metaphysea] region of long bones and within the
`cortical confinements of flat and short bones (Fig. 1}.
`Trabecular or cancellous bone is continuous with the
`inner surface of the cortical shell and presents a three
`dimensional lattice composed ofplates and columns of
`bone.
`The mechanical properties of canoellous bone have
`been studied less thoroughly than those of cortical
`bone. Tables 1 and 2 summarize the data of elastic
`modulus and compressive strength mentioned by
`different authors for different regions. These values
`originally expressed in different units
`(kg cm",
`kg Him—2, p inch—1, kN m”) were all converted to
`10‘ Nm". Comments refer to the condition of the
`specimen and the loading rate. Topography among
`and within bones is an important variable to be
`considered for the determination ofdensity, trabecular
`contiguity and mechanical properties of oancellous
`bone.
`This study concerns the mechanical behaviour of
`cancellous bone at
`the upper
`femoral
`region.
`Discussion therefore will be mainly restricted to data
`in the literature concerning this region. Cancellous
`bone of the upper femur has already been subjected to
`mechanical testing by Hardinge {1949}. Knese (1958}.
`Evans and King {1961} and Schoenfeld er of. {1974]
`(cf. Tables 1 and 2}. Slow strain rates applied by these
`authors, the use of osteoarthritic femoral heads ob-
`tained at the time of arthroplasty by Schoenfeld er at.
`or embalmed bone by Evans and King impede the
`interpretatioa of their results. The actual condition of
`the test specimens is not stated by Knese. This author
`
`Received 10 December 1932; in revised form 5 May £983.
`Supported by grant number 30071.76 Fonds voor
`Geneeskundig Wetenschappeiijk Onderzoelt.
`NE‘.
`|r|rl'.’—B
`
`97]
`
`Page 3 0f 15
`
`also tested only one specimen from each region of one
`bone.
`Exact knowledge ofmechanical behaviour ofcancel-
`lous bone at the different locations within different
`bone structures is valuable for the understanding of
`mechanical functioning of bone structures and it gives
`better insight into the particular architecture of bone
`tissue and bony elements at different levels of organiz-
`ation. Furthermore skeletal fixation of implants in
`joint replacement or in fracture treatment may depend
`upon mechanical behaviour of cancellous bone.
`Particularly the importance of structural and rnech»
`anical characteristics of cancellous bone at the upper
`femoral region for internal fixation of femoral neck
`fractures has been clearly demonstrated by the authors
`(Van Audekercke er al, 1979; Martens er of. 1979}.
`Strength of internal fixation and stability of this
`fracture depend to a great extent upon mechanical
`behaviour of the cancellous bone at the various parts
`of the upper femoral region. Failure ofcancellous bone
`at
`the femoral neck and intertrochanteric region
`precedes failure of the internal fixation device. So one
`can state that to some extent the ultimate properties of
`the cancellous bone and not of the foreign material are
`the limiting factors for the strength of internal fixation
`of femoral neck fractures. Sound principles of internal
`fixation of femoral neck fractures have to be based
`upon insight in architecture and mechanical behaviour
`of oanoellous bone at the upper end of the femur.
`
`MdTERIALS AND METHODS
`
`Source. preservation and selection of material
`Femoral bone specimens were obtained from 20
`autopsy subjects. The postmortem sampling was per-
`formed between 12 and 24 hr after death. The speci«
`mens were labeled and stored in a freezer at —- 20°C
`before testing. Data on age. serr and medical history
`were obtained. Standard anterioposterior and lateral
`
`Page 3 of 15
`
`

`

`9T2
`
`M. MARTENS. R. VAN AUDEKERCKE. P. DELPDRT. P. DE MEES'TER and J. C. MLILIEI
`
`
`
`uuoEa:322%.5.52%53E9...5862:2.0
`
`
`
`
`
`
`.382:003.on2:95mun—5.39.:fiEuEO
`
`33:3.3man35:9383?21.“3:2.53m35
`
`
`.Tfiuwxgod“E:35N.EEw:EU335:“in
`
`30.62%«not?5Eumans—Eon“.3minwas...
`
`
`
`
`EuEHoonm£359.;mafiafinw:£ufl63
`
`
`
`nmfi8.nun—Ewan.
`
`
`
`
`
`SE».55...::23uofifiafim
`
`«FE2.8E:SE»...
`
`
`
`
`
`B3;bin—Em“on:nfifioam.85:650
`
`EnuEEoU
`
`“553%$3
`
`
`
`own._033....on25:2:0:0“5560mm"a?
`
`
`
`
`
`
`
`E:2.2733min9.3newace—Enema.«Ehuczau
`
`“ion"Sofiaon.hewas5.2
`
`
`
`BEER—m3.33in.
`
`2%5:52.8332m
`
`32.53%5%ran._.3:BE52$3m3..Earn«
`
`ER2...50th59:5up;5::5.5925
`
`
`
`.5?EBet—«E505?
`
`wag—EBEENEE—5
`
`mcuEfion—w55
`
`
`
`
`
`mono:385a."mES.“unuEMuonmon
`
`33.23::ES;32m
`
`
`
`
`
`mum»:3.550..335330va
`
`2+
`
`:Hhv+
`
`:1.
`
`9:flH
`
`.Gw
`
`3...”
`
`nd
`
`ed
`
`dd
`
`«.5
`
`to
`
`.D.m
`
`and._.ha9mm._H“mm.
`
`.mG3mm
`
`mm
`
`hm
`
`am
`
`Km
`
`5*.
`
`.5
`
`own
`
`31m
`
`9mm
`
`5.6—2
`
`Lam
`
`v.5530:09.8homan—uni.m..”Bu...
`
`
`
`
`
`
`r.EZno:3—365m
`
`
`$5862....._0.3555SEEonm
`
`.553...
`
`
`
`Am:23:8EOE».15.3a:as;ESE...
`
`
`
`
`
`asnausea”Esau.“332
`
`
`
`9.252%_m:__u=:m=2
`
`553v3965.:
`
`sou:2.:.33:35".
`
`acumen—925
`
`
`
`“awn;HEaEum
`
`53:833¢a$18a$13
`
`
`
`
`
`“363$33.9.“3:5.—
`
`
`
`“25:25.3380.2x8:Eon—u..—A:
`
`
`
`
`
`
`
`:32
`
`m3—
`
`:32
`
`33
`
`06m
`
`“.3.
`
`0.3
`
`02.
`
`”.5W;
`
`
`
`
`
`gm.8.8.".3325’
`
`a:92335..
`
`_
`
`2
`
`
`
`Mumuon3523.?
`
`
`
`“863SEEP—m
`
`8.3.3
`
`mu_mE
`
`“295..
`
`$3....
`
`so:32%35F
`
`Sad2a..535
`
`Easr.3
`
`:3:was.—ucm"cam
`
`3mm:Busch
`
`Ema:83!
`
`53:3&5“.2
`
`6.39.33
`
`9.3Swap—855a..—
`
`
`
`3.3:.232
`
`
`
`38:2.85..
`
`:3_.m2“2..
`
`
`
`Uca.2ka
`
`Page 4 of 15
`
`.J
`
`Page 4 of 15
`
`

`

`
`
`Fig. l. A longitudinal section at the proximal part ofa femoral bone specimen shows the thick wall ofcortical
`bone tissue at the shaft gradually tapering offal the metaphyscal region. At the femoral head the cortical bone
`represents only a thin shell continuous with the underlying trabecular bone network. This trabecular network
`already shows at the macroscopieal level a varying porosity. A cartesian coordinate system is superimposed
`on the femoral neck for specimen orientation.
`
`
`
`"liq—nr-v--
`
`Page 5 0f 15
`
`9?3
`
`Page 5 of 15
`
`

`

`HYDRAULIC CYLINDER
`
`'14
`I'll/1
`l/l/l/I/l/l/l/m W/I/l/l/I-W/l/fl'
`
`
`
`MICRDMETER-SCREW
`
`DISPLACEMENT-LIMITER (R)
`EXT ENSOMETER 1'. Bl
`
`SPECIMEN
`
`LOAD CELL [Fl
`
`
`
`
`
`
`STEEL BALL
`
`Fig, 2(a) Diagrammatie drawing of the testing equipment. (b) Close up View of lrabecular bone specimen
`placed between two flat platens. The downwards displacement with a predctcnninated excursion ofthe upper
`platens results in a controlled deformation of the test specimen.
`
`974
`
`Page 6 0f 15
`
`Page 6 of 15
`
`

`

`
`
`
`
`Fig. 6. An arrangement of canoellous bone into three sets oflamellae can be observed at the upper femoral
`region. The sheets are lined up along the direction of these lameilae and the interconnecting struts can also
`been seen.
`
`WS
`
`Page 7 of 15
`
`Page 7 of 15
`
`

`

`
`
`Fig. 7. A scanning electron microsecpic view ofIhe cancellous bone at. the upper femur also reveals the sham
`and strut arrangement of the cancellous frame work.
`
`9T6
`
`Page 8 0f 15
`
`Page 8 of 15
`
`

`

`The mechanical characteristics of cancelleus bone
`
`97'!
`
`
`
`
`
`5.82....gas...5..30¢2.5EB.—..uEBonm255:0“$25...in...
`
`
`
`Wm.Jun:.33.32:".
`
`BEEEoUr.EZwe.x.35.0on
`
`
`
`suwflohmulmmEQEoU
`
`BEESQ32mR.i90.3
`
`
`
`3:3“5:mcufisonm.653.235.2.5:02.“.5...».33925
`
`
`
`
`
`33E9532maa.
`
`25ESE32m
`
`2EEEG32m
`
`bum—no.5...“
`
`
`
`33....E3.8.0.8.55533
`
`
`
`3H3Eflan..353
`
`5mb.on.595manna—«U
`
`andb."on.96“sound—an...
`
`
`
`
`
`mmwfimodsauna.mE:33:
`
`an563.9,2
`
`an8.35%.:
`
`
`
`
`
`oudHfi...SEES:.235
`
`
`
`
`
`nméflandx8:.EoEuu
`
`23:8
`
`
`
`
`
`
`
`
`
`En:a...382.30.5208.“..6.mugs::35.Wm”0.ad.mannin—REE.
`
`BOP—ME
`
`
`
`Eofiimnofifiamino_.3.2a..£2.382:3.....L3.2.9a.
`SS333an?!»adhe_«55.53“5:333:22.203mm3nd3.Ed
`
`domm3.3:
`
`Nd“fl8.3.5..
`
`
`
`._-EESE3.9SE5.33.5.332m$5:.mEiEm
`
`«.0ENBEE».
`
`3525:.
`
`We3.mu.“EH
`
`
`
`
`
`
`25:9:32.3.5nanitsmcufimuonm.030.0We30.9A..m.2m.—58%
`.D.m:36m3:.mEEELF3:mean:was3:5
`
`
`
`
`
`
`
`
`
`“60.33ham—25mweEa...“5&6on3.3.9.”..20dd53:.”3.3.“.3.3.5..»63:ESE...
`
`
`
`
`
`
`
`
`
`
`
`new5.5:..283.2.c235,on”canMEEERGmm“«.mHSqEvan3.52.5.2SR:.3a353E£2
`
`
`
`
`
`var—2.5:.3:399.50%.3“3.5.23...ca3an0.3.8....3&3
`
`
`asp—memu959wEnE.9.050.022..33.2
`
`
`2E22a32m3H2.Em2”.2%;EEO.82.«Spam
`
`
`
`SE52632moméflmowarcane»En....223
`
`anon38.33mm.“H3....v8:Eon—um.32.was.new39m
`
`
`
`
`
`
`
`2.:Essa3Q.—wmd5..cm.553.3...352.5}
`
`
`
`anon—.86.»sworn.nae..ionhug:«0.anEsme—u.»Goa:£351.30Us.».5533
`
`
`
`
`
`
`
`295.8%$3m...madam...”“v.02..Epozu;v.._o.383:.3.wSufism»
`
`
`
`
`
`
`
`“Emu;33E»...—.mmm:unocz
`
`
`
`
`
`
`
`
`
`matauESE».—mufli.233.2.62.33309m.2mmduse...EOE»...33:.__uaENEBEM
`
`
`
`
`
`
`
`
`
`
`
`.EoEea...E«5:80..55....».6.3:...»:32mm2in.$338.2089....32.:..ua32:»...
`
`
`
`
`
`
`
`
`
`.8...=<
`
`
`
`
`
`3.8qu“o...Eugooam.o.3957.cm.i9.12.8:8....8325}Ram:$9.9}
`
`
`
`
`
`
`
`Page 9 0f 15
`
`Page 9 of 15
`
`

`

`9?“
`
`M. Minn-ens. It. VAN Auceaeectts. 1’. Dacron. P. DE MEESTER and J. C. MLILIER
`
`X-ray views of the bone specimens were taken. Further
`X-ray examination of the specimens was performed on
`the bone slices after cutting. Based upon this screening
`several bone specimens were discarded from the group
`because of possible local alteration in the bone struc-
`ture by metastasis or pathophysiological osteoporosis.
`Specimen preparation
`
`The proximal part of the femur Comprising the
`epiphysis and metaphyseal region is mounted in a
`polyurethane foam block in a standardized manner
`allowing for predetermined cutting planes. Slices of
`bone were sawn along the XY. X2 and Y2 plane ofa
`Cartesian coordinate system superimposed on the
`femoral neck {Fig 1} allowing for specimen orien-
`tation. The upper and lower surfaces of these slices
`Were polished in a deep frozen condition using liquid
`nitrogen irrigation on a high speed milling machine.
`Cylindrical test Specimens were removed at predeter-
`minated and standardized location by a hollow saw of
`8 mm diameter.
`Damage of the test specimens was avoided by
`machining the bone samples under liquid nitrogen
`irrigation and the use ofa high speed milling machine.
`The orientation of the cylindrical specimens is per-
`pendicular to the plane of the bone slices. The finished
`test specimens were labeled and stored in a physiolog-
`ical solution before testing. A bone mineral analysis of
`the trabecular bone specimens was obtained by the
`photon absorption method using a Cameron Bone
`Mineral Analyzer. Height of the specimen was re-
`corded using a micrometer. Dry weight of some
`specimens was obtained after testing. Therefore the
`bone specimens were kept in an ether-ethanol solution
`for two weeks and dried at room temperature for one
`week.
`
`Testing machines and instrumentation
`The wet
`trabecular bone specimens were placed
`between flat platens and a downward displacement of
`the upper platens produced compression loading
`[Fig 2a and b). lnterposition of metal rings ofadjust-
`able height {R} allowed for a predeterminated defor-
`mation of 15'}“ of the length of the bone specimens
`during the compression test. A high strain rate [7 s' l]
`during testing has been accomplished by the use of an
`hydraulic cylinder. Displacement was recorded by a
`linear extensomcter (B) and force by a load cell (F ).
`Correct samples and useful data were obtained from
`189 test specimens taken from 6 femora and loaded
`along the X axis. 306 test specimens from 7 femora and
`loaded along the Yards and 24 test specimens removed
`from 2 femora for testing along the Z axis.
`Data reduction. The elastic modulus (E) can be
`calculated from the slope of the force displacement
`curve in the elastic region (Figs 3. 4 and S].
`l
`£=RE whereR=tyet
`i = length of the specimen
`A = cross section of the specimen.
`
`Maximum stress (a max] for compression loading is
`given by F maxM where F max is the maximum
`recorded force during compression of the specimen.
`Bulk specimen density is obtained by dividing dry
`bone weight of the specimen by its total volume. Bone
`mineral content of the specimen per cm is obtained
`directly from the Bone Mineral Analyzer.
`Compression tests of whole proximal femoral struc—
`tures. In order to gain complementary information on
`the role of trahecular bone in the mechanical proper-
`ties of the proximal femur two pairs of femora were
`tested in compression. Through a cortical drill hole at
`the lateral femoral cortex the cancellous bone at the
`center of the femoral head. neck and intertrochanteric
`region was removed by a hollow drill under radiosco-
`pica] control. The trabecular network at these regions
`was destroyed by this procedure leaving intact only the
`periphery and the cortical shell. The mate ofeach pair
`of bones was
`left
`intact. The distal part of the
`specimens was imbedded at the shaft of the femoral
`bone in epoxy [plastic filler P 38]. The specimens were
`loaded in compression with 15" inclination of the shaft
`from the vertical in order to imitate the direction of the
`resultant of the joint reaction force for walking. A fast
`loading rate was applied (2 cm displacement 5").
`Strength and stiffness of the structure could be ob-
`tained from the recorded force vs displacement curves.
`The failure mode was recorded by an image intensifier
`connected to a video tape.
`
`RESULTS AND DISCUSSION
`
`Figures 3. 4 and 5 are representative load deforma-
`tion curves for specimens with high~ medium and low
`compressive strength. Mean and standard deviation
`for elastic modulus and compressive strength of trabe—
`cular bone at the different regions of the upper end of
`the femur are listed in Tables 3—5. The three tables
`contain the values for compressive strength and elastic
`modulus of trabecular bone specimens for different
`directions of loading.
`
`Variation between bone specimens
`The variation of these two mechanical parameters
`between femoral bone specimens is high (Tables 3. 4
`and 5). Also the variation of mechanical properties
`within a bone structure between the three distin-
`guished areas (femoral head. neck and intertrochan-
`tcric region] differs from one bone to another. The
`pronounced individual variation between bone struc‘
`tures masks to a great extent ageing and anisotropy of
`trabecular bone.
`
`Anisotropy
`Townsend et oi. [1975] demonstrated anisotropy of
`the stiffness of mncellous bone at the human patella.
`The authors stated that this anisotropy is a function of
`position and is related to the changing orientation of
`the basic structural unit namely a sheet and strut
`model.
`
`Page 10 of 15
`
`Page 10 of 15
`
`

`

`The mechanical characteristics of cancellous bone
`
`979
`
`FIN)
`
`SPECIMEN 23I3IE
`
`700
`
`500
`
`l—
`_
`P‘S'N'
`500-
`
`{.00-
`
`30C!
`
`200*
`
`mL
`
`
`‘\‘
`
`R=lgfl:ll.L:105Nlm
`E:R.l_
`A
`
`PM
`
`E ..o‘
`
`_ Dlx10‘3ml
`
`L.
`1
`
`
` Ir-Blmr.
`
`F 'l N}
`
`500
`
`SPECIMEN zama
`
`:2:ng = 5.92105 mm
`E:RLfl
`
`
`500'-
`
`FIN}
`
`SPECIMEN 23I7IF
`R = Igor :11}. x105 Mm
`=R‘—
`A
`
`E
`
`in
`
`
`
`Figs 3. 4 and 5. Load defamation curves ofdifferem specimens. amax and E are obtained from the graphs.
`
`Page 11 of 15
`
`Page 11 of 15
`
`

`

`980
`
`M. Mattress. R. VAN AUDEKERCKE. P. Dmonr. P. DE Msesrcs and J. C. Mouse
`
`Table 3. Longitudinal axis of trabecular specimens along the x axis (cf. Fig. 2)
`
`
`Elastic modulus {E}
`Compressive strength
`Age and
`Region
`[a WHO“ N m‘ 1
`10" N m" 3'
`
`Specimen
`sex
`I: = number of samples
`Mean1S.D.
`Mean1SD.
`
`29
`
`37
`
`39 R
`
`23
`
`41
`
`28
`
`All spec.
`along x
`axis
`
`2'? M
`
`29 M
`
`32 F
`
`4'1 M
`
`56 M
`
`T? M
`
`Mean age
`45 311'
`
`Head in = 10
`Neck n = 5
`lntertrochanteric n = 14
`Head :1 = [(1
`Neck :1 = 9
`Intertrochanteric n = 13
`Head n = to
`Neck :1 = '1
`lnterlrochanteric n = [6
`Head 11 = 14
`Neck n = 3
`Intertrochanteric it = 16
`Head n = 15
`Neck n = ‘1
`lntertrochnnteric n = 11
`Head :1 = 10
`Neck 1: = 5
`Intertrochanteric n = 14
`Region
`N = number of specim.
`N = 6
`
`5.9122
`3.6112
`2613.4
`9.9132
`6.4148
`5.818
`1614.6
`18.7111
`1115.1
`11118.9
`6311.1
`2.413
`5115
`0.91118
`110.8
`5,912.2
`3.6112
`26134
`
`4691206
`2961153
`1631194
`1291265
`$011410
`5171663
`22481533
`202411113
`4091405
`10911414
`4971673
`2621412
`3891265
`84189
`64182
`4691206
`2961153
`1631194
`
`9001710
`9314.5
`616110?
`6616.3
`
`3.612.} 2631111)
`
`The particular organization and orientation of the
`trabecuiar network at the upper femoral region in-
`duoes anisotropy but anisotropy cannot be evaluated
`fully by the values of Tables 3—5 although the bone
`specimens are classified according to the direction of
`
`loading. The high variation between bone specimens
`interferes with the effect 01' anisotropy. The striking
`difference however in mechanical characteristics be-
`tween left and right femur of one pair of bones
`(number 39) with trabecular bone specimens of one
`
`Table 4. Longitudinal axis of specimens along the y axis (cf. Fig. 2]
`
`Elastic modulus
`Compressive strength
`Age and
`Region
`10" N m“
`lflr’Nm"
`Specimen
`sex
`n = number of samples
`Mean1S,D.
`an15D.
`
`
`39 L
`
`72
`
`11
`
`ID?
`
`57
`
`21
`
`9
`
`32 F
`
`49 M
`
`5'? F
`
`65 M
`
`66 M
`
`1'2 M
`
`81) F
`
`All spec.
`along y
`axis
`
`Mean age
`60 yr
`
`Head 1: = 16
`Neck it 2 2
`Intertrochanteric n = 19
`Head n = 1'?
`Neck 1’! = 13
`lntertrochanteric n = 22
`Head 1: = 13
`Neck 1: = 111
`Intertrochanteric n = 15
`Head 11 = 17
`Neck n = 13
`Intertrochnnleric n = 21
`Head 1: = 16
`Neck 1: = 1
`Intertrochanteric n = 23
`Head 11 = 1'?
`Neck II = 8
`Intertrochanteric n = 16
`Head :1 = 21
`Neck rt = 6
`Intertrochanteric n = 211
`Region
`N = number of
`specimens
`N = '1'
`
`12515.2
`1310.4
`4.612.?
`12.3112
`4.2114
`29115
`814.9
`3.111.?
`2812.2
`9316.4
`1812.6
`1311.9
`15.61113
`0.9
`6.715
`8613.6
`1.8109
`2116
`4.9126
`1812.4
`3913.8
`
`6431325
`71.41313
`2681190
`1411311
`2121133
`1611118
`6591734
`2461209
`2201351
`5961653
`2111130
`2091168
`214011485
`6‘1
`9101185
`6121410
`1291100
`1151102
`228121 '1
`2141264
`2691358
`
`8111604
`102113
`174184
`2.8113
`
`3311.5 3171293
`
`Page 12 of 15
`
`Page 12 of 15
`
`

`

`The mechanical characteristics of cancellous bone
`
`93]
`
`Table 5. Longitudinal axis of specimens along the z axis (of, Fig. 2)
`
`
`Elastic modulus
`Compressive strength
`Age and
`Region
`[0" N m"2
`10‘ N m‘2
`
`specimen
`sex
`a = number of samples
`MeantSD.
`anfSD.
`
`2?
`
`16
`
`All spec.
`along :
`axis
`
`65 M
`
`"F0 F
`
`Mean age
`67.5 yr
`
`Head n = 6
`Neck n = 3
`lntertrochanteric n = 3
`Head :1 = 5
`Neck H = i
`Intertrochanteric n = 6
`Region
`N = number of
`specrmens
`N .—_ 2
`
`5.8}43
`1230.3
`0.4591109
`4.12
`0.?3
`03430.26
`
`351.1220
`63,369
`"1.9! 1.4
`4503155
`58
`1116.4
`
`403.5,I'65.Tfi
`4.9312?
`63;“10?
`(talisman
`
`0.595.111.205 11456.43
`
`bone being tested along the X axis and the specimens
`of the mate along the Yaxis demonstrates clearly the
`anisotropy between these planes. Hardinge (1949}
`noticed that the average force required to crush the
`cancellous bone at the femoral head was almost equal
`in corresponding zones for left and right samples. The
`difference in mechaniml behaviour for lelt and right in
`our experiment can be accounted for by anisotropy.
`Furthermore one notices from the data of this pair of
`bones and also from the mean values listed at the
`bottom of the three tables that
`the difference in
`strength and compressive modulus between the X. Y
`and Z axis vary for the three regions (head, neck.
`intertrochanterit: region). The highest discrepancy for
`these mechanical parameters according to the different
`axis of loading is seen at the femoral neck where the
`elastic modulus and also compressive strength along
`the Yand Z axis is only a fraction ofthe value for elastic
`modulus and strength along the X axis. Although the
`two specimens with trabecular bone samples tested
`along the Z axis are from an older age group one may
`conclude that the mncellous bone is weakest when
`compressed along the Z axis. This is especially true for
`the intertrochanteric region.
`The differences in anisotropy between these three
`regions reflect the different organisation of the trait:-
`cular pattern at
`these distinct areas of the upper
`femoral region. The dependence of mechanical be-
`haviour of cancellous bone on its particular frame
`work is further elucidated in the discussion of mech-
`anical properties and density.
`
`Variation within a specimen
`
`With regard to regional variation within a femoral
`bone specimen for compressive strength and elastic
`modulus one may conclude from the data presented in
`Tables 3—5 that for any loading direction the highest
`values are found at the femoral head. However, the
`magnitude of difference between the values at the
`different regions is dependent upon the loading direc-
`tion. For compression along the X axis the discrepancy
`
`between the E and a max of trabecular bone at the
`femoral head with regard to the E and a max for the
`femoral neck is significantly less than for a loading
`direction along the Yor Z axis.
`Trabecular bone of the intertrochanteric region is
`stronger and stiffer than the spongy bone at the
`femoral neck for the Y axis of loading but
`it
`is
`significantly weaker for the X and Z axis of loading.
`This relates to the unequal anisotropic behaviour of
`the various regions. These data are in contrast with
`Evans and King(196l) who found that specimens from
`the femoral head had the greatest compressive strength
`followed in descending order by specimens from the
`neck and greater trochanter whereas the modulus of
`elasticity according to these authors was highest in the
`neck followed by specimens from the head and finally
`the greater trochanter {one specimen only). We assume
`that the limited number of specimens tested by these
`authors is responsible for their contrasting findings.
`Finally. the standard deviation for the two studied
`parameters is extremely high at the different regions
`illustrating the particular distribution and organiz-
`ation of trabccular bone at the upper femoral bone.
`Analysis of the data (cf. Tables 3—5) reveals a lower
`dispersion of the values of compressive strength for the
`femoral head specimens compared with the femoral
`neck or the intertrochanteric region. Hardinge {1949),
`who determined the force required to crush the
`cancellous bone at the head and subcapital region.
`noted that
`the cancellous bone with the greatest
`compressive strength was related to the orientation of
`the compression lamellae extending from the medial
`end of the inferior cortex of the neck to the middle of
`the superior aspect of the head. The strength is greatest
`where the tension and compression lamellae are inter-
`secting. Our observations are in agreement with these
`findings.
`The contradictory results of Schocnfeld et at. (1914}
`who concluded that the relativer weak and strong
`regions of cancellous bone in the femoral head appear
`to occur on a random basis, can be explained by the
`
`Page 13 of 15
`
`Page 13 of 15
`
`

`

`932
`
`M. Marcus. R. VAN Aubexettctte. P. Daron. P. De Mess-res and J. C. MULIER
`
`they used osteoarthritic heads removed
`that
`fact
`during surgery for hip joint replacement.
`
`Ageing
`
`Because of the high variation between bone speci-
`mens any conclusion with regard to age changes for
`these number of tested specimens cannot be made.
`Ageing effects are masked by high individual dif-
`ferences between femoral bones.
`
`Correlation between compressive strength and elastic
`modulus
`
`Correlation between compressive strength and
`elastic modulus yielded high correlation coefficients
`for the specimens in the three loading directions.
`
`Specimens loaded along X axis in = 189} R = 0.84
`Specimens loaded along l’axis (n = 306] R =0.85
`Specimens loaded along Z axis (a = 24] R = 0.80.
`
`Therefore we can conclude that stiffness is closely
`related to strength in the use of cancellous bone.
`
`Correlation between mechanical properties and density
`Several authors have shovvn a definite relationship
`between mechanical properties ofcancellous bone and
`density. Weaver and Chalmers {1966}. Galante or al.
`{1970), Behrens et al. [1974), Schoenfeld et al. (1934]
`have shown high correlations between density and
`compressive strength or elastic modulus.
`Density differs considerably between bones result-
`ing in a high dispersion of mechanical behaviour.
`Density of trabecuiar bone also varies, within a
`femoral bone (Fig. 6]. Different concepts can be
`applied to determine density oftrabecular bone such as
`real density. apparent density. bulk density, ash weight,
`percent porosity. bone mineral content of a specimen
`per cm (B.M.C.l. We chose the last method as a
`measurement of density and for some bones we also
`determined bulk density. Bulk density is defined as dry
`weight of the specimen divided by total sample volume.
`
`A significant correlation could be shown for most bone
`specimens (cf. Table 6]. The correlation between
`B.M.C. and rt max is in most instances higher than the
`correlation between B.M.C.. and E. The specimens are
`grouped in Table 6 according to the axis of loading
`along the X and Ydirection. The correlation between
`mechanical properties and B.M.C. i