View Full Version : Tesing and fixation of bones

10-04-1992, 11:52 PM

On 28th August 1992, I offered Moshe some advice regarding his
proposed testing of rat bones and I note that he has decided to use
a bending test with the PMMA potting.

I guess it would have been helpful to provide my comments to the
rest of the network; however, I thought that most of these items
were well understood in the biomechanics community - or, perhaps,
that's why Moshe obtained only a few replies. Consequently, I
include my original comments below and provide some additional


As with most bones, you will find that the mechanical properties
are dependent upon:

+ Location of the sample.
+ Species.
+ Age.
+ Exercise.
+ Individual tested.
+ Other possible factors.

Destructive tests (comparing a treatment with some series of
"controls"), therefore, are fraught with the statistical challenges
of these biological variations. Any studies that cast aside
concern for these variables are likely to produce such large
standard deviations that no meaningful differences can be

Ultrasound (as a non-destructive tool) offers great advantages for
such monitoring; however, an "error analysis" should be conducted
(using the known dimensions etc and their associated measurement
errors), to ensure that the measurement tolerances will not
overwhelm the anticipated differences that are to be measured. The
size of the rat bone may well be a dominant factor in whether you
can expect useful data and the Error Analysis will advise you of
this problem BEFORE you start to sacrifice animals and time.

The non-uniform shape and anisotropic properties of long-bones tend
to frustrate the repeatability of bending tests; consequently, many
researchers have resorted to torsional tests. You may still wish
to use a guide-wire (surgical k-wire) to establish a meaningful (?)
axis for the bone.

While some researchers use low-temperature bismuth alloys for
potting their samples, these materials can be quite expensive.
Medical researchers often use PMMA due to its availability in the
hospital - material is often available free in the hospital, after
expiry of a "shelf life" or the opening of a larger pack that has
not been completely used. Bulk purchases of PMMA from Dental
Suppliers may be cheaper than the material used for total joint
replacements. Due to the large amount of heat given off by curing
(polymerizing) PMMA, the volume of PMMA should be kept to a
minimum. 5-minute epoxy is often readily available to non-medical
workers and is relatively cheap. Any of these techniques are
suitable for potting, so long as attention is paid to minimizing
potential thermal damage to the specimen.


I note that Moshe has decided to use a Bending Test with an Instron
machine and has encountered loosening problems with the samples
potted in PMMA.

INSTRON: It is unclear whether the chosen machine is screw-driven
or electro-hydraulically controlled. Due to the smallness of the
samples (rat bone), a screw-driven machine could present additional
problems in obtaining smooth curves. No matter whether the machine
is screw-driven or electro-hydraulically controlled, the resolution
of the position-measurement and force-measurement transducers will
be of particular concern.

BENDING TEST: It is still unclear whether the proposed tests are
destructive (with all of the inherent problems listed above) or
whether the same sample is to retested within its "elastic" range
after a variety of treatments.

A bending test will require some mechanism for controlling the
axis-of-bending in relation to the principal axes (i.e.
geometrically-defined - second moment of area) of each bone.
Attempts to force bending to occur on any other axis than the
"minimum" principal axis may well result in the bone trying, during
loading, to rotate on the loading and support noses.

ASTM provides guidelines (D790M-86 standard on plastics) for span-
to-depth ratios for bending tests on machined samples of materials.
For anisotropic materials, span-to-depth ratios of 60:1 are
recommended! Where samples may be of various dimensions, the
failure to control "shear effects" (by adhering to an appropriate
span-to-depth ratio) may create additional problems and increase
the variance in the data.

ASTM-D790M-86 also provides guidelines for the radii of the loading
and support "noses", "... to avoid excessive indentation, or
failure due to stress concentration directly under the loading nose
or noses..."

ERROR ANALYSES: An Error Analysis should be conducted BEFORE any
type of testing and is not just a requirement for the Ultrasound
Testing mentioned above.

DEHYDRATION: Moshe does not mention any precautions that he is
taking to control the extent of dehydration of the bone.

The curing PMMA alone will generate heat that can dehydrate the
bone. The significance of such dehydration is not widely
published. In a recent paper:

Finlay et al, 1992: "Embalming effects upon the mechanical
properties of bone: Preliminary experiments" In "Experimental
Mechanics: Technology transfer between High Tech engineering and
biomechanics", Little, E.G. editor, Elsevier Science Publishers
B.V., Amsterdam, 185-200,

(Details of the proceedings of this conference were provided on
BIOMCH-L on 19th September 1992)

we reported briefly upon the dramatic anisotropic contractions that
can be seen when the cortical bone of a bovine femur is dehydrated.
The contractions were 0.6% axially, 2% circumferentially, and 3%
radially and were statistically different from one another. These
values translate to strains of 6,000, 20,000 and 30,000 microstrain
respectively - i.e. considerably above the 3,000 microstrain peaks
that are reported in the literature for normal activities of daily
living! While embalming reduced these strains, they still
remained at levels of 0.5%, 1.2% and 2.1% respectively (these are
all approximate figures without quotations of the standard

Rehydration of the bovine samples in our tests brought the axial
dimensions back close to their original values; however, the
circumferential and radial dimensions remained notably different
from their original dimensions.

These findings for the "contractions due to dehydration" validate
observations quoted to me by Dr Sidney Lees of the Forsyth Dental
Center in Boston. (Sidney is not on e-mail). Clearly this
information on "dehydration/ rehydration" strains needs to be
disseminated more widely - any editors out there who would like to
review such a manuscript for their journal?

Obviously, any contraction effects due to dehydration of the rat
bone will add to the contractions encountered during polymerization
of the PMMA and will produced a loosely-potted sample.

Careful choice of a bismuth-alloy will provide an alloy that
maintains its dimensions during cooling; however, it will be more
expensive than PMMA. The volume required for testing rat bones
should, however, be quite small.

To aid in the alignment of long-bones during potting, we regularly
use screws (3 or 4) in each pot. These screws and each pot can be
coated with Vaseline (or some other "thin" grease), to permit easy
removal and reassembly of each bone in a single set of pots. In
this way multiple sets of pots are not required. The screws may be
utilized to "take up" any "small" amounts of "play" between the
bone and its pot; however, we have not encountered this "looseness"
problem with bones from humans, cows or sheep.

I hope these additional comments are of value and I would encourage
other BIOMCH-L members to post their comments for general viewing -
obviously, I was in error by not displaying my original comments to

Best wishes:

Bryan Finlay, PhD 519-663-3063
Director of Orthopaedic Research 519-663-3904 FAX
University Hospital
P.O. Box 5339
London, Ontario, CANADA, N6A 5A5