. lllllUJllllllllllllllllllllllllllllllHlNllNlllllHllllllll ; \/
312930

1088 3696

LIBRARY
Michigan State
University

—

 

 

 

 

This is to certify that the
thesis entitled
The Effects of Rehydration Time and Anatomical Source
on the Mechanical Response of Cancellous Bone

Allografts Used in Cervical Arthrodesis

presented by
Kathleen Marie Cowling

has been accepted towards fulfillment
of the requirements for

' -
Master 3 degree in Biomechanics

 

 

Kilgifiob 4' me,

Major professor

Date ’7 ,/é/<? :7

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

(“5‘4"
l.

. 5}

 

 

MSU

LIBRARIES
.—::—.

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

S

' ~,
h‘
.1 s
, r
;. :x’ i"
‘. \.-

(a 3"in ‘5‘

in

 

 

 

THE EFFECTS OF REHYDRATION TIME AND ANATOMICAL SOURCE
ON THE MECHANICAL RESPONSE OF CANCELLOUS BONE
ALLOGRAFTS USED IN CERVICAL ARTHRODESIS
BY

Kathleen Marie Cowling

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE
Department of Biomechanics

I 987

ABSTRACT

THE EFFECTS OF REHYDRATION TIME AND ANATOMICAL SOURCE
ON THE MECHANICAL RESPONSE OF CANCELLOUS BONE
ALLOGRAFTS USED IN CERVICAL ARTHRODESIS

BY

Kathleen Marie Cowling

The purpose of this research was to investigate the mechanical
response of cancellous bone allograf ts. These bone dowels, used in the
Cloward technique of cervical arthrodesis, were tested fresh and
rehydrated for various time intervals after lyophilization to determine
which time of rehydration restored the grafts mechanical properties to
those of the fresh bone. Comparisons of anatomical source sites were
done to discover if there are differences between the sites where the
allogenic bone was taken from the donor.

Results from these tests indicate that differences were found
between the donor's right/ left leg, medial/lateral condyie, and f lexion
contact points. These differences were in the mechanical response after
rehydration for longer than two hours. Within each donor, transverse

isotropy was seen at dominant sites and was sensitive to the rehydration
effects.

ll

ACKNOWLEDGEMENTS

The author wishes to thank some special people who helped her by

providing support while this work was being conducted:

To Robert w. Soutas-Little for his encouragement and serving in the
patient role of major advisor.

To Joia Stapleton Mukherjee who jointly conducted research on a

parallel study and cheerfully shared laboratory space.

To Herbert Reynolds, Richard Borman, and Nicholas Altiero for serving

as invaluable resources by objectively reviewing the work before
print.

To Robert Hubbard for initiating interest in this project and for his
valuable assistance. ‘

To the Michigan Regional Tissue Bank at Edward w. Sparrow hospital
in Lansing, for providing the monetary support for this study as well
as the source of the testing material.

And to her family, who survived the frustrated moments and never
doubted her ultimate success.

iii

TABLE OF CONTENTS

Page O
LIST OF FIGURES ............................................................................ v
LIST OF TABLES ............................................................................ vii
Section
I. INTRODUCTION ............................................................................ I
ll. SURVEY OF LITERATURE ........................................................... 8
III. MATERIALS AND METHODS ...................................................... l3
IV. RESULTS ............................................................................ 25
V. DISCUSSION ............................................................................ 32
BIBLIOGRAPHY ........................................................................... 45

iv

LIST OF FIGURES

Figure Page
i. Different techniques developed for cervical
arthrodesis ................................. 2
2. The Cloward technique ....................... 4

3. Proposed neck injury references for axial
compressive neck force, for adult population. . . 5

4. Distal femur showing site locations, and set
up for hole saw with 1.4 cm bit ............... l4

5. Typical allograf t approximately i.4 cm X 2.0 cm
shown here scaled next to a standard metric
ruler ....................................... 16

6. Experimental specimen matrix ................ l7

7. Allograf t as seen end on before compression
between lucite grips, shown here 8/7 actual
size ........................................ l9

8. End of allograf t before testing as seen through a
10x microscope ............................... 20

9. Testing apparatus, showing Instron Materials
testing machine, lucite grips, Nicolet oscillo-
scope and Wild microscope with camera
attachment .................................. 21

i0. Screen of Nicolet oscilloscope showing typical
stoke and load curves recorded in volts/seconds. 22

1 1. Typical Load vs. Deformation curve showing 2%
offset method used for obtaining peak load and
stiffness .................................... 24

12. Peak load vs. ash weight density ............... 26

13. Linear regression analysis of normalized peak
load vs. donor age ............................ 31

14. Illustration of distal femur showing directions
of applied physiological and testing loads ...... 33

15. Line graph of normalized peak load values for
donors 1, 2, and 3 ............................ 34

16. Line graph of normalized peak load values for
donors 4, 5, and 6 ............................ 36

i7. Line graph of normalized stiffness values for
donors 1, 2, and 3 ............................ 37

18. Line graph of normalized stiffness values for
donors 4, S, and 6 ............................ 38

19. Hexogonal models of rod-like and plate-like
columnar structure of cancellous bone .......... 39

20. Schematic representation of the sample sites
as they relate to the condylar surface .......... 42

21. Comparison of sites from the medial condyle. . . .43

22. Comparison of sites from the lateral condyle. . . .44

vi

LIST OF TABLES

Table Page
1. Group 1 data ................................. 27
2. Group 2 data ................................. 28

3. T-test comparing means for normalized load at
different rehydration times to fresh control,

both sides, all sites, all donors of each matrix
included ..................................... 29

VII

I. INTRODUCTION

Arthrodesis (artificial fusion) of the cervical spine is used to
treat both traumatic and nontraumatic conditions of the spine, including
idiopathic cervical spondylosis, vertebral fracture, and f lexion-extension
cervical spine injuries. The objective of the procedure is to remove all
degrees of freedom from the joint so bone ingrowth can permanently fuse
the vertebrae together. The most common reasons stabilization via fusion
is desired are: osteoarthritis with nerve root irritation, disc
degeneration, and unstable fractures. 54 Various anterior approach
techniques have been developed by three separate groups in the late
1950's; Smith and Robinson 1953 55, Cloward I958 22, Dereymaker
1965 23. The anterior approach made it possible to not only provide
stabilization and decompression by fusion of the vertebral bodies but also
to obtain access for removal of symptomatic pathological lesions
previously inoperable. 22 The difference between each of these
techniques is basically whether the operation is limited to simple
discectomy and interbody fusion or if entrance into the spinal cord to
remove osteophytes is desired, 45 Figure 1.

The Cloward technique, which uses a cylindrical bone dowel,
permits direct access to the spinal cord before the graft is put in place.
Unfortunately, this creates the possibility for iatrogenic spinal cord
injury due to improper use of the instruments or posterior displacement of
the graft either during surgery or postoperatively. However, this

technique also permits the removal of lesions and complete decompression
1

loletol

 

  

  

fir

llll

 
    

' 7’;
I Ill'.1

I.
I

”I ll

Type 3 . Aniuoposieiiot

         
    

 

FIGURE I. Different techniques developed for cervical
arthrodesis. Type 1, Robinson and Smith, Type 2,
Cloward, Type 3, Bailey and Badgely. White 62

3
of the nerves, occasionally a necessity. Rectangular grafts have a firmer
fixation than dowel grafts, but do not always immobilize the joint from
rotation.

The operation is performed with the patient lying supine with the
cervical spine extended. The surgeon enters anteriorly, Figure 2 A, and
obtains direct access to the fusion site using retractors, Figure 2 B. .The
vertebrae involved are separated using spreaders, and a hole is made
between the superior and inferior vertebra using a drill, Figure 2 C. All
fragments are removed and the disc is cleared away. The graft is seated
in place and the spreaders removed, Figure 2 D. The soft tissue retracters
are removed with subsequent closing. Following implantation, the graft
must bear the weight of the head and superior vertebrae as well as
muscular loads.

The graft must sustain its structural function for the duration of
the bone induction . Schneider recorded 85% of patients having complete
fusion within four months. 54 During that time the grafts primary
purpose is to maintain the disc space so nerve impingement does not
occur, and until union is complete, to serve as an inorganic supportive
matrix for new bone growth. White and Hirsch found that autogenic grafts
maintain integrity under loads up to 2 1/2 times body weight. Mertz 44
found that injury could occur under compressive neck loads of 4000 N for 0
msec and only 1 l 10 N applied for 30 msec, Figure 3. However, serious
neck injuries can occur from several loading conditions including bending
or torsional moments and tension forces. 1 l 10 N was considered to be
the minimum load the transplanted grafts should withstand so as not to be
a weak link while the joint is still unfused.

Autogenic bone is considered to be ideal for grafts because no

 

FIGURE 2. The Cloward technique, A; the anterior approach,
B; exposure of the vertebral bodies, C; hole drilled to size,
to accept corresponding graft, D; graft seated in place,
and spreaders removed.

Axial Compressive Neck Load

Newtons

8000

6000

4000

2000

 

3 o 50
msec o__._____io vo . o 4

Duration of loading over given load level

FIGURE 3. Proposed neck injury references for axial
compressive neck force, for adult population. Mertz 44

6

antigenicity is involved and viable cells are thought to survive and speed
healing. in actuality, only osteocytes within 300 u of the surface survive,
and the rest die from lack of vascularization. 7 The major objections to
the use of autogenic bone are that a second invasion is required to retrieve
the bone, and often the volume or quality of the bone desired, is not
available from the patient. Consequently, if performed, the procedure will
increase morbidity and make recuperation longer for the patient.

Since the beginning of the United States Navy Tissue Bank in
1950, many tissue grafts have been processed for use in reconstructive
surgery. From the start, research interest has focused on the field of
bone grafting because it is the easiest tissue to transplant, and the need
for deriving an equivalent replacement for autologous bone is high. 7

Canceilous bone for preparation of allografts used by tissue banks
can be harvested from several sites, including: the distal femur, femoral
head, proximal tibia and vertebrae. The demand for intact femoral heads,
and the difficulty of retrieval from vertebrae often preclude those sites
from being utilized. The distal femur and proximal tibia are left as good
consistent sources for obtaining several grafts from the same donor. Due
to the geometry of its planar contact surface, the proximal tibia is also a
source for cancellous blocks, commonly used in reconstruction; therefore,
the distal femur is the most likely source of dowel grafts.

There are several different ways allograf ts can be processed,
varying both storage and sterilization: 2 lyophilization (freezing at -70' C
while dehydrating in vacuum so final water content is less than 5%) or
storage at -70' C and kept fresh. Both methods greatly decrease the
antigenicity of the Done by killing the viable cells which might be viewed

as immunogens by the recipient‘s body. Both can be sterilized using Co 601

7
radiation, but only lyophilized bone can be sterilized using ethylene oxide
gas. Due to ease of processing, shelf life, and shipping considerations,
the preferred method is to lyophilize and irradiate the bone.

This study concentrated on the lyophilized grafts and attempted to
determine the rehydration time required to return their material
properties to those of fresh bone. Also, it was important to determine
which site(s) of the distal femur supplied the maximum strength in
resistence to compression, that is, whether from a dominant leg, condyle
or position as specified with degree of f lexion.

A parallel study investigated the effects of the irradiation
sterilization process on the bone and constructed a theoretical model

using finite element method to compare the experimental results to the
analytical ones. 47

II. SURVEY OF LITERATURE

One of medicine's oldest needs has been to be able to replace
diseased or damaged parts of the body. For awhile, this was limited to
using artificial limbs or dentures, but in 1878, MacEwan 40 performed
the first successful bone transplant from one person to another. In 1895,
an introduction to bone grafting techniques was published by Barth. 3 Then
in 1937,0rell 48 began using cadaver bone for graft material. The
demand for grafts increased dramatically during the Korean conflict and
prompted the Navy to develop the first large scale collection and
preservation facility. Since 1950, the US. Navy Tissue Bank has taken
over 37,000 tissue deposits, and dispensed two-thirds of the grafts made
for clinical use. 7

Since Barth's first introduction of bone grafting, many
applications have developed including: arthrodesis of joints, filling bone
cavities after infection, replacing joint surfaces or bone lost to trauma or
tumor removal. The two primary forms of material used in these
procedures are autogenic (person's own) bone and allogenic (taken from a
donor) bone.

The anterior approach to the cervical spine was first described by
Chipault 20 in a textbook of Neurosurgery published in 1895. This
specific approach was not used again until the 1950's when three groups
independently developed it for the use in spinal fusion. Smith and

Robinson 55 designed a horizontally seated graft using rectangular block.
8

9

Bailey and Badgely 2 used rectangular block also but positioned vertically,
providing the option of fusing more than one joint with one graft. Cloward
22 developed a cylindrical dowel graft, which was limited to single joint
fusion as was Smith and Robinson. Recently, in i984, Gore 3' created a
new technique using a dovetail design with an odd shaped rectangular graft
which is seated vertically. Like that of Bailey and Badgely, it can be used
for multi-level fusions, adding to the options available to orthopaedic
surgeons.

Increased use of bone grafts turned attention to the preparation of
the tissue. The first attempt at preserving bone using refrigeration for
later transplant was done in 1912 by Carrel. '7 Shortly afterwards,
lnclan 34 tested the viability of frozen human bone for use in orth0paedic
surgery.

Flosdorf 27 developed the processing method of lyophilization, and
bone was the first tissue lyophilized and used clinically. Schneider 54
reported a success rate of 92% in 57 patients of cervical arthrodesis using
lyophilized allograf ts. 85% of the unions demonstrated bridging within 4
months, as examined radiographically. Kline 37 found that freeze-dried
allograf ts induce the recipient‘s mesenchymal cells to differentiate into
mature osteoblasts aiding in the incorporation of the graft. Kreuz 38
reported no difference in the remodelling and repair of lyophilized bone
when compared to fresh autogeneous bone, except that the rate was
slower. Carr and Hyatt ‘6 using histology to confirm these results.
Brown 9 showed no radiologically demonstrable significant difference in
the process of incorporation between autograf ts and allograf ts. Pelker
49 determined lyophilization did not have a deleterious effect on the

compressive strength of bone allograf ts. However, Triantafyllou 59

IO

found a considerable reduction in the mechanical properties of the bone
allograf ts, due to lyophilization and sterilization by irradiation.

Burwell '2 in 1963 listed several criteria for using allograf ts:
the bone must be sterile, nonantigenic, capable of induction and
resorption, permit vascularization, and finally be implanted close to
cancellous bone.

According to Elves 25’ lyophilized grafts did not stimulate
humoral immunity and produced a decreased cell-mediated response
compared to that of fresh bone. A year later, in 1977, Bright 7 stated
that the freeze-drying process kills the viable cells in the graft,
subsequently decreasing its antigenicity. Bright 6 found that rehydration
of 24 hours restored the lyophilized cortical bone so that its mechanical
properties were closest to those of fresh bone.

Studies done by several different groups and reported by
Galante 29 indicated that cancellous bone acts as an anisotropic material
resulting in different strengths depending on the direction in which the
load is applied. Gibson 30 determined a dependence of the mechanical
properties of cancellous bone on density, cell wall properties and cell
geometry. She also stated that the symmetry of the cancellous structure
is dependent on the direction of the applied loads , and the density was
dependent upon magnitude of those loads. Galante 29'. Weaver and
Chalmers, 6° and Behrens, 4 found a strong linear relationship between
compressive strength and density. Martens, 42 and Lindahl, 39 found the
correlation to be 0.7 between strength and density.

Behrens 4 also showed that bone strength varied with position in
the joint . They chose to differentiate the sites using the contact points

at 0', 45', and 90’ of flexion as their specified sample sources. They

l i
found the bone from 0' and 45' of the lateral condyle and 45' of the medial
condyle to be the strongest sites. They ranked the sites with 45' being
the strongest, 0' slightly lower, and 90' the weakest. They determined
the medial condyle to be stronger than the lateral one (all sites from each
condyle considered collectively). These results suggest not only
magnitude but duration and frequency of the applied loads on the bone site
are the determining factors of bone strength.

Medawar 43- in 1944, investigated the immunologic basis for
tissue rejection and, found autogenic bone generally was a superior graft
material. Chalmers, '9 in 1959, showed bone to have the same
histocompatibility antigens as skin, but, in 1979, Richards 53 discovered
that bone cells have a smaller number of antigens than skin.

Duthie 24 in 1958, tested the rates of vascularization of cortical
and cancellous bone, and found the cancellous had a faster rate, confirming
the earlier work of Abbott I in 1947.

Lindahl 39 discovered a quantitative deterioration in most of the
strength parameters of bone with age. The change in strength was greater
than the decrease in density (increased porosity) could explain, leaving the
possibility that there is a qualitative change as well. Specifically, that
trabecular reorganization may be responsible for the change. Weaver and
Chalmers 50 found that after the age of forty there is an faster, rate of
decrease in strength as aging continues. Bright 5 hypothesized that the
changes in aging bone might be due to increased cross-linking of the bone
collagen. Weaver and Chalmers also noticed that trabecular
reorganization might explain the variations in bone strength which occur
independently of changes in density. Behrens 4 suggested this possibility
too, and considered it to be a significant factor. Therefore, strength is

l2
dependent on several factors which must all be considered together.

This study examined the effects that anatomical site and
processing have on the mechanical properties of cancellous bone
allografts. Specifically, whether lyophilized grafts could respond
similar to fresh bone, and if so, what rehydration time would be required.
The site of the bone is an important question from the viewpoint of .the

tissue bank, since the findings could change the choice of sites utilized
for load bearing grafts.

III. MATERIALS AND METHODS

Tissue samples of cancellous bone used for testing were supplied
by the Michigan Regional Tissue Bank, a division of Edward W. Sparrow
Hospital in Lansing Michigan. Samples were obtained using the Tissue
Bank's standard procedure of procurement and then stored at -70'C until
processed. When thawed, they were cleaned of any remaining soft tissue
and cut according to predetermined guidelines using a Black and Decker
hole saw with an attached 1.4 cm bit. The cortical bone plates on the end
were removed with a Stryker autopsy saw, so the samples were
composed of cancellous bone only.

Since donor variability was a concern, consideration was given to
maximize the number of allograf ts obtained from one donor. This study
was restricted to One sex, male, and an age range of 34 to 55. This
removed the sources of scatter due to decreasing bone density caused
either by aging, disease, or the onset of menopause in women. A
repeatable procedure was desirable so that all donors could be processed
the same, and differences due to anatomical site could be analyzed. For
consistency in protocol, both right and left distal femurs from the same
donor were prepared as follows; six allograf ts were drilled from each
side with three coming from each condyle. The grafts were numbered
according to the condyle and postion at approximately equal spacing,
from anterior to posterior of the articulating surface, Figure 4. The
dowels, taken from the medial condyle were numbered 1, 2, and 3, with

13

 

FIGURE 4. Distal femur showing site locations,
and set up for hole saw with 1.4 cm bit.

l5

site 1 being the most posterior, and sites 2 and 3 then taken from
increasingly anterior positions, ultimately dividing the articulating
surface into thirds. The same was done to the lateral condyle using the
numbers 4, 5, and 6, with 4 taken as the most posterior. Each prepared
graft was identified by donor, leg side (right/ left) , condyle

(medial/ lateral), and corresponding site. All dowels were taken from
these designated sites and varied only in length, a result of the differing
condylar sizes. The lengths ranged from 9.5 mm to 25 mm with the typical
graft being around 20 mm, Figure 5.

The grafts were sorted according to experimental design, Figure
6. This specimen matrix was used to insure that all variables were
considered equitably. Donor, side, condyle, site, and treatment were all
treated as possible influences on the responses observed. The dowels
were then stored fresh in refrigeration or ly0philized to reduce the water
content to less than 5% of total composition. 27 The fresh grafts were
tested within 48 hours of drilling and the lyophilized ones were
rehydrated in sterile saline the specified time. Rehydration times ranged
from 2 to 24 hours with 6, 12, and 18 hours covering the intermediate
range.

Since the major load on the vertebrae, in VII/0, is that of axial
compression, the grafts were tested under compression using
diametrically opposed loading as applied by grips geometrically analagous
to the inferior and superior vertebrae simulating the in viva conditions.
Deformation tests were run to a maximum value of 30 it diametral strain
corresponding to the loss of disc space. Also of concern was the mode of
failure, since brittle fracture would produce loose fragments which could
cause damage in the spinal canal.

 

FIGURE 5. Typical allograf t approximately 1.4 cm X 2.0 cm
shown here scaled next to a standard metric ruler.

I7

TREATMENT

I 2 3 4 '5 6

 

SITE ( anterior! posterior position and condyle)

 

l mom 1 2 3 ' 4 s 6
E3, 1 LEFT 2 3 4 s s l
g 2 mom 3 4 s s l 2
m
g 2 LEFT 4 s 6 l 2 3
8 3 mom 5 6 l 2 3 4
3 LEFT l 6 l 2 3 4 5

TREATMENT CODE I=FRESH, 2=2 HOURS, 3=6 HOURS, 4= I 2 HOURS,
5' I 8 HOURS, 6=24 HOURS

FIGURE 6. Experimental specimen matrix. Treatment number corresponds
to different lengths of rehydration time or the fresh control.

18

Test fixtures were designed to simulate the loading as would
occur postoperatively. The grip compression surfaces were made to
geometric specifications to approximate the superior and inferior
vertebrae as prepared surgically, Figure 7. The grips had compression
surfaces with concave rounds of radius 7 mm running the length of each.
When the test sample was positioned before testing, a gap between grips
of 4 mm would approximate the same distance as the desired disc
spacing found in Vii/0.

Before testing, the grafts' lengths were measured with a
micrometer so that density values could be determined using the
volumetric measurements. Since Galante 29 had reported the effects of
trabecular orientation during testing, all grafts were aligned before
testing with the apparent trabecular grain running vertically, Figure 8.
Even though the primary axis of loading would be parallel with the long
axis of the cylinder, the effects of the radial trabecular structure were
considered potentially significant.

Tests were performed using an Instron Servohydraulic Materials
Testing Machine with stroke control at 3%lsec strain rate to a maximum
30 73 strain , Figure 9. Both load and stroke data were recorded
using a Nicolet oscilloscope and stored using an attached mini-floppy disk
drive. A DEC PDP 1 1/23 minicomputer was used to transfer the data
from the floppy disks and plot load / deformation curves on a PRINTRONIX
p-series printer, Figure 10. This was done using four programs which
changed the initial data recorded as volts/seconds into values of
Newtons/millimeters while removing the time base. The first program
moved the data from the floppy disks recorded by the Nicolet and
transferred it to the hard disk on the PDP. The second program then

 

FIGURE 7. Allograf t as seen end on before compression
between lucite grips, shown here 8/7 actual size.

20

 

FIGURE 8. End of allograft before testing as
seen through a 10X microscope.

21

 

FIGURE 9. Testing apparatus, showing lnstron materials
testing machine, lucite grips, Nicolet oscilloscope and
Wild microscope with camera attachment.

22

 

FIGURE 10. Screen of Nicolet oscilloscope showing typical
stroke and load curves recorded in volts/ seconds.

23
converted the values which are recorded in pounds per inch into Newtons
per millimeter. The third program removed the time base and paired the
load and stoke data point to point. Finally, the fourth program plotted the
values calculated by the previous program onto a pair of axes.

The load vs. deformation curves were analyzed using a 2% offset
method to determine the peak load and stiffness values. The stiffness was
defined as the slope of the linear portion, line 1, Figure 1 1. A second
line was then drawn, parallel to the first but offset to the right on the
deformation axis, line 2. Then peak load, line 3, and deformation, line 4
were the values determined from where lines 3 and 4 intersect with the
curve. Stiffness was calculated from the peak load and deformation
values. Those values were then normalized by the mineral densities
obtained from ashing. Linear regression analysis was used to calculate
the correlation between age, density and compressive strength. Line
graphs were constructed from the peak load and stiffness values obtained
for the different rehydration times and anatomical sites. A t-test was
done to compare the statistical means of the different treatments to that
of the fresh control. After testing, the grafts were ashed in a muffle
furnace at 850°C for 90 minutes to remove all organic material and water.
The grafts were then weighed on a Mettler analytical balance to calculate
percent mineral density. The ash densities were used as the
normalization parameter for peak load and stiffness, in an attempt to

remove some of the variance due to length and donor differences.

LOAD (N)

8500
6000
5500
5000
4500
4000
3500
3000

2500

2000 ‘

1500
I000

500

24

 

 

 

-. 1 2
.2 3......
_’ ’I ‘
[Iir'fi T'TTT TIIUTIIUTTT'TITT In TT'VTUT'TTITIWTT'TTT l ' TUTT'TT Til
0.0 1.0 2.0 3.0 4.0 5.0 6.0

DEF ORMATION' (mm)

FIGURE 1 1. Typical Load vs. Deformation curve showing 2%
offset method used for obtaining peak load and stiffness.

IV RESULTS

Upon first examination the raw data showed considerable scatter,
possibly attributable to the fact that the prior health of each individual
donor varied greatly and could not be controlled for in this study. Previous
researchers have used ash weight density as a normalization parameter
for compressive strength of cancellous bone. The first analysis made was
a comparison of compressive strength and ash weight density using the
method of linear regression, Figure 12. The results were very similar to
the values reported by Weaver 60 for vertebral and calcaneal cancellous
bone. The correlation coefficient determined for N-52 of all sites tested
fresh was 0.79, and all data were then normalized by ash weight density.

The analysis involved examination of the potential changes in
response of graftsfrom each donor, left and right legs considered
independently, to different rehydration times. All sites within the knee
were assumed equitable, with each donor's grafts separated according to
experimental design. Two groups, each comprised of three donors were
tested after four different treatments. Grafts from both groups were
tested fresh and at two hours rehydration. The first group was then tested
at 6 and 18 hours, while the second group was then tested at 12, and 24
hours, Tables 1 and 2. The normalized loads for each graft of a particular
treatment were averaged, and a t-test was done to compare their means
to that value determined for the control of each group, which was the
fresh bone, Table 3 . The results of this test revealed that for the first

group there was no significant difference between treatments. However,
25

26

of

h

Euxu CLmzmo Ema} xm<
mfi
Idll

o.h

 

528 Ems; Se .0, 82 been fi mane...

mN ON

mud n L

own— lxmcg n»

 

coo—
CON—
cov—
com—
com—
OOON
CONN
OOVN
oomN
oomN
coon
CONN
oovn
comm
comm
ooov
OONV
OOVV

N UVOT )IVEId

27

TABLE 1. Group 1 data. D=donor, Sl=side, ST=site, TX=treatment, ASH=ash
weight, LOAD=peak load, LENGTH=length, AX=deformation, DENSITY=
density, NLOAD=n0rmalized load, and NSLOPE=normalized stiffness.

081 ST TX ASH

d—MUNuN—e—NUN—N—Uuu-‘NNU—

C‘UAMNNC‘UIA-‘UDJNA-‘UIG-‘NUTAOU

OOOOOUIUIUTUIUIUINNNNNNUIUIMUDIU

9

1.14
0.96
1.06
0.60
0.93
1.30
0.90
1.06
1.03
0.99
0.53
1.09
0.75
0.83
0.67
1.13
0.88
1.18
1.04
1.00
1.07
1.06
0.50

LOAD LENGTH

N

2600
3050
2650
1180
2600
3900
3000
1600
3700
1300
600
4400
1750
1700
1750
1550
1900
1900
2700
3050
2600
2600
1050

mm

18.37
14.48
16.18
14.47
18.35
19.46
17.27
15.06
17.26
13.33
9.48

15.04
14.45
14.47
16.10
18.31
19.47
18.34
18.32
14.40
15.51
19.40
1 1.10

AX DENSITY NLOAD NSLOPE

mm

0.80
0.70
0.70
0.60
0.72
0.77
0.55
0.52
0.65
0.50
0.50
0.82
0.60
0.57
0.52
0.40
0.60
0.50
0.54
0.60
0.57
0.65
0.47

g/nrzl

4.03
4.31
4.26
2.69
3.29
4.34
3.39
4.57
3.88
4.83
3.63
4.70
3.37
3.73
2.70
4.01

2.94 '

4.18
3.69
4.51
4.48
3.55
2.93

N/b

x 101

645.2
707.7
622. 1
438.7
790.3
898.6
885.0
350. 1
953.6
269.2
165.3
936.2
519.3
455.8
648. 1
386.5
646.3
454.5
73 1.7
676.3
580.4
732.4
358.4

N/amm
x lo'

806.5
101 1.0
888.7
731.2
1097.6
1 167.0
1609.1
673.7
1467.1
538.4
330.6
1 139.3
865.5
799.7
1246.4
966.3
1077.2
909.0
1355.0
1 127.2
1018.3
1 126.8
762.5

TABLE 2. Group 2 data. D-donor, Sl-side, ST=site, TX=treatment, ASH=ash

weight, LOAD=peak load, LENGTH=length, AX=deformation, DENSITY=
density, NLOAD-normalized load, and NSLOPE-normalized stiffness.

0 SI ST TX

UIU'IAACIUIUIACIO‘AAAO‘CIU'IU'IOIOUIUIAAAOIOIUIUIA

GUAM—0‘5NUU‘IC‘UAUN—OUAUN—OUAUN—

—-—-———NNNNNNLALAAbOOOOOOC‘C‘O‘O‘C‘O‘

ASH

9

0.82
0.96
0.80
0.85
0.97
1.00
0.97
0.97
0.98
0.86
0.67
0.77
1 .01
1.09
0.59
0.90
0.84
0.82
0.90
0.54
0.86
0.86
0.97
0.50
1 . 12
0.89
0.94
0.92
0.74

LOAD
N

2075
2725
2325
1700
2975
3200
3050
3000
3050
2050
1550
2040
2800
3275
875
2150
2300
1675
2000
875
1500
1450
2475
1500
1450
1300
800
975
1450

LENGTH AX
mm mm
18.3 1 0.65
1 8.33 0.53
16. 1 2 0.65
l 7.25 0.60
1 8.35 0.60
1 8.35 0.70
1 9.50 0.65
1 8.40 0.70
16. I 0 0.65
17.20 0.52
1 8.30 0.50
1 8.30 0.50
19.48 0.58
20.00 0.63
1 8.32 0.60
1 9.48 0.60
1 7.27 0.65
I 8.36 0.50
20.06 0.58
17.34 0.48 -
19.42 0.55 .
20.02 0.48
1 7.24 1 .00
1 4.43 0.68
17.06 0.60
21 .45 0.55
1 9.41 0.50
20.99 0.67
I 4.00 0.60

2.91
3.41
3.23
3.19
3.43
3.53
3.23
3.43
3.95
3.25
2.38
2.73
3.37
3.54
2.09
3.00
3.15
2.90
2.91
2.03
2.88
2.79
3.66
2.26
4.28
2.69
3.15
2.85
3.43

Mb

x10I

713. 1
799.1
719.8
532.9
867.3
906.5
944.3
874.6
772.2
630.8
651.3
747.3
830.9
925. 1
418.7
716.7
730.2

' 577.6

687.3
43 1.0
520.8
519.7
676.2
663.7
338.8
483.3
254.0
342. 1
422.7

DENSITY NLOAD NSLOPE
g/nr2l

N/amm
x10I

1097.1
1507.7
1 107.4
888.2
1445.5
1295.0
1452.8
1249.4
1 188.0
1213.1
1302.6
1494.6
1432.6
1468.4
697.8
1 194.5
1 123.4
1 155.2
1 185.0
897.9
946.9
1082.7
676.2
976.0
564.7
878.7
508.0
510.6
704.5

29

TABLE 3. T-test comparing means for normalized load at
different rehydration times to fresh control, both sides,
all sites, all donors of each matrix included.

TREATMENT NLOAD MEAN .t sd N p value
N/g/cm3 sample size

MATRIX I ( 3 DONORS)

FRESH 615811567 5 -----
2 HOURS 593.213687 6 >040
6 HOURS 683.8: 157.1 6 >025
18 HOURS 518411082 6 >010
MATRIX 2 (3 DONORS)
FRESH 770. 1 3 122.7 6 -----
30 MINS g 368.2: 87.8 5 <0.0005
2 HOURS 583.01 106.8 6 (0.01
12 HOURS 699911806 6 >020

24 HOURS 765.5: 134.0 ' 6 >040

30

the second group had significance at both the 30 minute and 2 hour
treatments. The difference for the 30 minute interval can be disregarded
because, grafts tested at this time failed in a brittle mode. The other
finding possibly indicates that although 2 hours is adequate to remove the
brittle failure, it may not be restoring the bone completely back to the
level of fresh bone.

An analysis was made to compare age and compressive strength
using the method of linear regression. Eleven donors ranging from 34 to
55 had six grafts, one from each site, tested fresh for each. The average
of those six sites was then used to determine the linear regression line
and correlation coefficient, Figure 13. When all i i donors were included
the correlation coefficient was -0.78. This line agrees with the literature
from Lindahl and Weaver, in that a decrease in strength is seen after age
40, and continues to decline. Lindahl also found that the decrease in
strength with age, was greater than the decrease in density could explain,
implying that there must also be a qualitative change in the structure as
well. Without further research, including ages covering the range more
evenly, it is difficult to conclude what the actual trend would be,
specifically where the line begins to decrease most rapidly.

0f the total 193 grafts tested, 8.8% failed to meet the minimum
criterion of sustaining load up to 1 l 10 N. These grafts had no traits in
common which could explain this weakness. They came from all
treatments, several donors of different ages, lengths, and densities. This
finding would indicate the need to examine the trabecular structure
closely to determine if any specific pattern would correlate with the
failing grafts.

Normalized load

Ncm3lgx10'

31

840
820 .
600
780
760
740
720
700
680
660
640
620
600
580
560
540
520

, 500

Y - -8.3 X + 1049
r- -0.78

 

 

 

3031323334353637383940414243444546474849505152535455
donorage

FIGURE 13. Linear regression analysis of normalized peak load vs. donor age.

V DISCUSSION

As well substantiated in the literature, cancellous bone is an
anisotropic material, having preferred axes aligned in the directions of
in VII/0 principal stresses. Bone grafts from the femoral condyles, Figure
14, are taken such that the cylindrical axis corresponds with the direction
of maximum compressive stress. The radial axes would then be in the
transverse plane to this principal stress. Milne-Thomson, 45 described
this particular type of anisotropy as transverse isotropy. With the
primary axis being the axis of monotropy, and all directions perpendicular
to the axis of monotropy being equivalent. Williams and Lewis, 64 also
commented on this particular case of orthotropy, after testing lyophilized
cancellous blocks from the tibial epiphysis, rehydrated overnight.

In a parallel study done by Mukherjee 47, 2 hours was determined
to be the minimum time of rehydration required for non- irradiated
lyophilized grafts to avoid brittle fracture failure. Based on this finding,
all strength and stiffness data were normalized additionally by the 2 hour
rehydration value for each donor. The results from the first group are
shown in, Figure 15. The most obvious difference is between legs
within any given donor. The strength of grafts taken from one leg of each
donor remains fairly constant with increased rehydration while the other
shows an increase of compressive stength after 2 hours up to the 6 hour
interval. The second group shows similar results with the exception of
one donor, number 5, who consistently had very high strength values for

grafts from both legs at all the treatments, Figure 16. The values for
32

33

MAXIMUM in viva COMPRESSION

 
   

 

 

TDISTAL

3 A
POSTERIOR

 

FIGURE 14. Illustration of distal femur showing directions of
applied physiological and testing loads.

PEAK LOAD NORMALIZED TO 2 HOURS

34

\ DONOR 1 LEFT

/\ DONOR 2 LEFT
...—-\—--<:"""" "" T"

\ oouon 3 R161"

4\ - DONOR l nToTlT
“—m-E-Z‘T DONOR 2 NIGHT

DONOR 3 LEFT

 

 

 

 

o2 4601012141618202224

1100115

FIGURE 15. Line graph of normalized peak load values
for donors 1, 2, and 3.

35
stiffness for group one were comparable to the trends of compressive
strength, Figure 17. However, the values for stiffness of donor 5 of
group two were also higher compared to that of the other two donors,
Figure 18. Donor 4 showed little changed in stiffness while donor 6
had similar results to that of the first group.

Williams described the different types of trabecular structure as
forming cylindrical arrangements in the primary vertical direction. Some
consist of interconnected rods forming circular or elliptical holes in
the transverse plane, while the other has curved plates running parallel to
theprimary axis. Gibson, 30 used a hexagonal model to represent the
different types both based on the cylindrical shape, but one having an open
framework, while the other more closed, Figure 19.

if the 2 hour minimium rehydration time sufficiently saturated the
structure enough to remove the chance of brittle fracture failure, but did
not completely fill the inner trabecular microstructure with fluid, then
when tested in compression along the minor axis, the plate-like structure
would respond different than the rod-like structure which would be less
effected by the enhanced hydraulic effect. However, with increased
reconstitution, the plate-like structure would demonstrate considerable
increase in compressive strength as the trabecular microstructure was
saturated giving the properties of fluid filled cylinders. Ultimately
responding with greater increase in strength than the rod-like structure,
which may have a dominant axis but which does not demonstrate the
anisotropy of the plate-like framework. This is supported by several
groups which have commented on the anisotropic design of cancellous
bone, but have not dealt with the problem of rehydration. As mentioned,
Williams rehydrated overnight, missing the period of time where the

PEAK LOAD NORMALIZED T0 2 HOURS

9.9.9
GIN:

36

/,,00ll0li 6 llloliT

// ' DONOR 4 LEFT

/ a, ‘ Rees; _. _ ., g 400110115 LEFT
,« " ‘ ~ ~ «0011011 5 illoll'r

co—nusmauaco

,/

\ \
\

\\\

\

l.

‘i‘

I

I"

I

{i

°--rrrrrrN

 

AA

024601012141610202224

1100115

FIGURE 16. Line graph of normalized peak load values
for donors 4, 5, and 6.

STIFFNESS NORMALIZED T0 2 HOURS

 

  

37

0011011 1 LEFT

 

/\_ 2..» bunch 2 LEFT
,— w w H“:.__ fig

/ 0011011 3 1110111'

 

“I“ 0011011 2 1110111

~

.5 ._ .— ~-—- ~""‘""'"‘T"'e 00N011 1 1110111

 

/ --———- -—--—a 0011011 3 LEFT

v——

024601012141610202224

1100113

FIGURE 17. Line graph of normalized stiffness values

for donors 1, 2, and 3.

STIFFNESS NORMALIZED TO 2 HOURS

°----—-‘-

la

lob£—I~:Li°.sbib'~i°c=

.<=.c.<=
CING

95:
AM

 

 

38

k
l
/ \x
\ .

‘ // N\\\

\ / \\

l / >00N011 5 1116111
l / , ___,____ 0011011611101”

‘ / / ,5; ”4% 0011011 5 LEF T

00N011 4 LEFT

 

00N011 41110111
0011011 6 LEFT

 

024681012141610202224

1100118

FIGURE 18. Line graph of normalized stiffness values
for donors 4, 5, and 6.

 

39

 

1%
d A Tb/F’Wgr

/ l— ______./—J

 

 

‘\

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FIGURE 19. Hexogonal models of rod-like and plate -like columnar structure
of cancellous bone. Gibson, L. J.. J. Blomechanics .vol. 18, no.5,
1985, p. 317.

40

transverse isotropic effects are the greatest.

It is consistent with previous studies that the dominant leg would
have a more plate-like trabecular microstructure due to higher load while
the less dominant would have a rod-like trabecular structure. Since none
of the cancellous bone grafts were loaded along their axis of monotropy,
the increase in strength and stiffness in the radial direction with
rehydration would be a measure of greater transverse isotropy and
therefore leg dominance. From this view, it is probable that the response
demonstrated by the different legs of a donor, is due primarily to the
involvement of rehydration and transverse isotropy. With the most
dramatic results seen in the dominant leg, where this particular structure
is most apparent. The exceptional donor from group two, as mentioned,
showed increased strength and stiffness, consistently throughout the
treatments, possibly due to an increased fitness level in that individual. A
very active donor would probably not show the difference between legs, as
would someone whose dominant leg carries the load for a greater
proportion of the time.

The analysis of strength dependency on site location yielded results
somewhat consistent with the literature. Wismans, 65 found that the
contact points with f lexure angle followed irregular patterns on the
condyle surfaces. A surface map showing these contact points is shown in
Figure 20. Comparison with Figure 4 shows that the samples sites did not
correspond with contact points of 0', 45', and 90', the locations used by
Behrens. 0n the lateral condyle most contact occurs between sites 4 and
5, while on the medial condyle sample sites correspond closer to contact
points. Using strength sensitivity to rehydration as a measure of

transverse isotropy and therefore strength, Figure 20, shows that site 2 or

41

approximately 45' would be the heaviest loaded in viva. This is
consistent with the results reported by Behrens and in Figure 21, but on
the lateral condyle, the transverse isotropy increases around the most
posterior site, site 4, Figure 22 . Behrens reported higher strengths
between 0' and 45' which would correspond to site 5. Three factors might
explain this discrepancy; the contact area for all flexure angles is '
concentrated in the area of both sites 4 and 5 and second, the results
presented here do not differentiate between right or left leg or age, and
finally the raw data shows similar transverse strength values for all
areas on the lateral condyle.

The question of proper rehydration time depends upon the
application. From the point of view of basic tissue biomechanics, strength
will depend upon rehydration time and type of trabecular structure
requiring care in comparison of results from different laboratories. For
the tissue banks, removing the possibility of brittle fracture failure is the
most important aspect and therefore 2 hours rehydration appears
adequate.

This thesis has presented a possible explanation for the change in
strength and stiffness with continued rehydration. By use of this
hypothesis, increase in strength or stiffness would indicate there would
be greater transverse isotropy and therefore higher in viva loading along
the axis of monotropy. To fully examine if this theory holds true, cubes of
the material should be taken and adjoining specimens tested along

different orthogonal axes. This will be left as a future research project.

42

.80— .RQD .6285 .o 69:23 .885 m._ cocoo—
bucoacmt 8.5 33:8 65 .6 Ewfima 389C .8226 .5328 05 S
328 .65 mm 6.3.5 0388 65 .8 8388858 Someocom .om mag;

Iv
._<mm._.<._

 

Ail
mo_mm._.mon_

PEAK LOAD NORMALIZED TO 2 HOURS

43

 

-_-_————-~
O-NUAMC‘NGOC

 

I
l

O
c:
I
I
I
l
I
1
\
I
I
I
in
z
to
u

99.
ON

 

0 2 4 6 01012141610 20 22 24

1100115

FIGURE 21. Comparison of sites from the medial condyle.

PEAK LOAD NORMALIZED T0 2 HOURS

44

‘
” \

NM 311.9 4

cL-Nusmclsiacc

.4-—-slte 6
09 \ flan-“‘3'...

0.0 “~—- r”"'""’
0.7
0.6

 

‘ 5111! 5

 

 

A

0 2 4 6 01012141610 20 22 24

1100115

FIGURE 22. Comparison of sites from the lateral condyle.

 

BIBLIOGRAPHY

BIBLIOGRAPHY

Abbott, L.C., E.R. Schottstaedt, J. 8. Dec. M. Saunders, F. c. 8031.
Evaluation of Cortical and Cancellous Bone as Grafting

Material. WM- '10- 2.
April 1947. p. 381.

Bailey R.W., CE. Badgely.Jnunna]_nLB_QiJe_ang_,J_ojnt_SnnganL. vol. 42,
1960, p.565.

Barth, A. Histologische Untersuchungen Uber Knochenimplantation
sttr, Eathn], Anat, 17:65 1895.

Behrens, J.C., P.S. Walker and H. Shoji. Variations in Strength and
Structure of Cancellous Bone at the Knee. a_B_j_o_me_cnan_i_cs_,
vol. 7, 1974, p. 201.

Bentley, 6., J.E. Aston. Transplantation of bone, cartilage and

synovial joints. Tissua Transplantation. chp. 14. Clinical
Surgery International vol. 3. Peter J. Morris ed. Churchill
Livingstone, New York, 1982.

Bright, Robert W., and Hans Burchardt, The Biomechanical properties

of preserved bone grafts. QstansnnngnaLAflggnafisnngjngx
Banking and Cliniga] Annligations. chp. 23. Little, Brown and

Company. Boston.

Bright, Robert W. ,Gary E. Friedlaender, Kennethw. Sell. Tissue
Banking: The United States Tissue Bank. W
vol. 141, no. 7, July 1977. p. 503.

Bright, Robert W., James D. Smarsh, and Vernon M. Gambill,
Sterilization of Human bone by irradiation. _Qs_tansnnn_d_r_al
5 10.. Soul 3. _|i.9 1 .j ‘00] 310 Chp,21,

45

10.

II.

12.

13.

I4.

15.

16.

17.

18.

46
Little, Brown and Company. Boston.

Brown, M. D., T. l. Malinin, and P. B. Davis. A roentenographic
evaluation of frozen allografts versus autograf ts in anterior
cervical spine fusions. 5110000520.. 1 19:231, 1976.

Burnstein, Albert H., John D, Currey , Victor H. Frankel, and Donald T.
Reilly. The Ultimate properties of bone tissue: The Effects of
Yielding. W. vol. 5, 1972, p. 35. '

Burwell, R.G., The fate of freeze-dried bone allograf ts.

lmnsolantatlomflmceedlnosvol. 8. no.2. Supplement 1. June
1976.

Burwell, R. 6. Studies in the transplantation of bone. V The capacity
of fresh and treated homograf ts of bone to evoke

transplantation immunity. Jnnrnal of Bans ang goint
M4SBz386 1963.

Burwell, R. 6., Studies in the transplantation of bone. V111.
Treated composite homograf t-autograf ts of cancellous bone:
an analysis of inductive mechanisms in bone transplantation.

Jaurnal 91 Bang and Jain; Snrgany, vol. 488, no.3 August
1966.

Bush, L. F. The Use of Homogenous Bone Grafts. alonrnal 0f Bans
anMintjnmam no. 3 July 1947. p.620.

Campbell, C. J., T. Brower, D.G. Macfadden, E. B. Payne, J. Doherty.
Experimental Study of the Fate of Bone Grafts. Jnurnal

9f Bonn and Joint Snrgany, no. 2. April 1953. p. 332.

Carr, C. R., G. W. Hyatt. Evaluation of Freeze-Dried Bone Grafts.
WW no. 3. June 1955. p. 549.

Carrel, A., The preservation of tissue and its application in Surgery.
JAMA_59, 1912 p. 523.

Carter, Dennis R., Wilson C. Hayes. The compressive behavior of bone
as a two phase porous structure. Wm

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

47
Sam. vol. 59A, no. 7, Oct, 1977.

Chalmers, John, James K. Weaver. Cancellous Bone: Its Strength and
Changes with Aging and an Evaluation of Some Methods for

Measuring Its Mineral Content. Jnurnal of Bone and 29101;
W. vol.48A, no. 2, March 1966.

Chipault, A. Chirurgie Operatoire du Systeme Nerveux Reuf f et Cire
Ed., paris, 1895.

Cloward, R. 8.. Gas-Sterilized Cadaver Bone Grafts for Spinal
Fusion Operations. Sninggol. 5 no. 1. Jan/Feb 1980.

Cloward, R. 8., The Anterior Approach for removal of Ruptured
Cervical Discs. WVOI. 15 , 1958, p. 602.

Dereymaker, A, J. P. Chosez, and R. Kennes. Les Traitment

Chirurgical de la Disc0pathie Cervicale. CUMLIDQL 40,
1965 p. 113.

Duthie, R. 8., A histochemical study of transplanted skeletal tissue

during tissue culture in viva. British anurnal nf Plastic
Sangeny . val, 11, 1958, p. 1.

Elves, M. W. Newer Knowledge of the immunology of bone and
cartilage. Clingntnnn, RalRas. 1202232, 1976.

Evanof f , Jean, Sterilizing and preserving human bone. 25080095111021
April 1983, vol. 37, no. 5.

Flosdorf, Earl W., George W. Hyatt, The preservation of bone grafts
by freeze-drying. Sunganuol. 31, no. 5, May 1952.

Friedlaender, Gary E., Henry J. Mankin, Transplantation of
osteochondral allograf ts. W vol. 35, 1984, p. 31 l.

Galante, J. , W. Rostoker, and R. 0. Ray, Physical Properties of
Tabecular Bone. W 5, 1970.

Gibson, L.J., The Mechanical Behavior of Cancellous Bone. Jaumatnt

31.

32.

33.

34.

35.

36.

37.

38.

39.

41.

48
WVOI. 18, no. 5, 1985.

Gore, Donald R., Technique of cervical interbody fusion. Clin, Qnthon.
B_eJ_,R_es_.no. 188, Sept. 1984.

Hart, Muriel M., Edward D. Campbell, Jr., and Mark G. Kartub, Bone
Banking: A Cost Effective Method for Estabilishing a

Community Hospital Bone Bank. Clin. Ortnnn. Bel. Basno.
206, May, 1986. p.295.

Hunter, Letha Y., Ethan M. Braunstein, and Robert W. Bailey,
Radiographic changes following anterior cervical fusion.
Sninavol. 5, no. 5. Sept/Oct. 1980.

lnclan, A The use of preserved bone grafts in orthopaedic surgery.
Jnnrnal at Bong and Jaint Sangany, WI. 24, 1942, p. 81.

Jonck, L. M. , Allogenic bone transplantation. SA Madiaal ganrnal, l 2
Sept. 1981.

Kaplan, Saul J., Wilson C. Hayes , John L. Stone, and Gary S. Beaupre.

Tensile Strength of Bovine Trabecular Bone.J. Binmachanics .
vol. 18, no. 9, 1985.

Kline, Stuart M, and Steven R. Rimer, Reconstruction of osseous
defects with f reeze-dried allogenic and autogenous bone.
Wmvolmé. Oct. 1983.

Kreuz, F. P. G. W. Hyatt, and T. C. Turner. W
MEL 35A, 1951, p. 863.

Lindahl, 010v, Mechanical Properties of Dried Def atted Spongy Bone.
We 47. 1976.

MacEwen, W. The osteogenic factors in the development and repair of

bone. W WI. 6, 1887, p. 289.

MacKenzie, AP, The physico-chemical basis for the f reeze-drying
process. International symposium on f reeze-drying of bio-
logical products. Wvol. 36, 1976 p. 51.

42.

43.

44.

45.

47.

48.

49.

50.

51.

52.

49

Martens, M., R. Van Audekerck, P. Delport, P. De Meester, and J. C.
Mufter. The mechanical characteristcs of cancellous bone at

the upper femoral region. Mimics, vol. 16, no. 12,
1983.

Medawar, P. B. The behaviour and fate of skin autograf ts and skin
homograf ts in rabbits. math vol. 78, 1944, p. 176.

Mertz, H. J., V. R. Hodgson, L. Murray Thomas, G.W. Nyquist, An
assessment of compressive neck loads under in jury-producing

conditions. WW. Nov. 1978.

Milne-Thomson, L. M. -Plan

Elasticity. MRC Technical Summary Report * 152, April 1960,
Madison, Wisconsin.

Mosby. Wings. Edmonson and Crenshaw
edits. 6th edition 1980. p. 1983.

Mukherjee, J. S.. ”The effects of sterilization and storage
procedures on the mechanical properties of bone allografts."
MS. Thesis, Michigan State University, 1987.

Orell, 5. Surgical bone grafting with " Os Purum,“ ”0s Novum" and

”boiled bone.” WWI/01 19. 1937
p.873.

Pelker Richard R., Gary E. Friedlaender, and Thomas C. Markham.

Biomechanical properties of bone allograf ts. Clin= Orthnn.
Rel. Res. 174, April 1983.

Prolo, Donald J., Sally A. Oklund, Sterilization of Bone by Chemicals.
teo h r ° 1 in n 11 i
Wchp. 22. Little, Brown and company. Boston.

Pugh, James W., Elastic and Viscoelastic properties of Trabecular
Bone: Dependence on structure. WVOI 6, i973.

Reilly, Donald T., Albert H. Burnstein, Victor H. Frankel, The Elastic
Modulus for bone. W 1974.

53.

54.

55.

56.

57.

58.

59.

60.

61.

50

Richards, R. R, F. Langer, P. Halloran, and A E. Gross. The antigenic
profile of the mouse skeletal system. QLLDQLQSILQ
lnansagtjnns. vol.3,1979,,p. 164.

Schneider, J. R., and R. W. Bright, Anterior cervical fusion using

preserved bone allograf ts. Inansnlantattnnflmsaeding; vol.
8, no. 2. supplement 1,June, 1976.

Smith, G. W., R. A. Robinson. The Treatment of certain cervical spine
disorders by anterior removal of the Intervertebral Disc and

lnterbody Fusion. MW vol. 40
l 958.

Sokal.R. R. FJ. 801111. Blomettlt. West

W. W. H. Freeman and Co. San
Francisco. 1967.

Stone, John L., Gary S. Beaupre, and Wilson C. Hayes. Multiaxial
Strength characteristics of Trabecular Bone. W101.
WI. 16 no. 9, 1983.

Townsend, P. R., P. Raux, R. M. Rose, R. E. Miegel, and E. L. Radin. The
Distribution and Anisotropy of the Stiffness of Cancellous

Bone in the Human Patella. Was. vol. 8, 1975, p.
363.

Triantafyllou, N., E. Sotiropoulos, and J.N. Triantafyllou, The
mechanical properties of the lyophylized and irradiated bone

grafts. WWI. 41, no. 1, 1975.

Weaver, James K. and John Chalmers, Cancellous Bone: Its strength
and changes with aging and as evaluation of some methods for

measuring its mineral content. Jannnal at Bang and Joint
MVOI 48A, no. 2, March 1966, p.289.

Welland, Andrew J. , Thomas W. Phillips and Mark A Randolph, Bone
Grafts: A radiologic , histologic, and biomechanical model
comparing autograf ts, allograi ts, and free vascularized bone

grafts. Wvol 74. no. 3. Sept.
1 984.

62.

63.

64.

65.

51
White, A. A. lll,et al. QUILQELDQL vol.91, 1973, p.21.

White, A. A. The role of stabilization in the Treatment of
Cervical Spine Injuries. Spins. vol. 9 no.5 1984 p.512.

Williams, J.L., J.L. Lewis. Properties and an Anisotropic Model of
Cancellous Bone From the Proximal Tibial Epiphysis.

WWI. 104 Feb. 1982.

Wismans, J., F. Veldpaus, and J. Janssen. A Three-dimensional

Mathematical Model of the Knee- joint. J, Biamechanics.
vol. 13, 1980, p. 677.

 

"11111111111111.111111111115