View Full Version : Preconditioning in Soft Tissue Test: Summary

08-25-2002, 10:31 PM
Dear colleagues,

Since I put the question on the net last week, I received three responses. I
would like to thank them for their constructive comments and discussions,
which really helped me to understand the problem.

Here is a short summary for our discussions.

John Z. Wu, Ph.D.
Research Scientist
National Institute for Occupational Safety and Health
Centers for Disease Control and Prevention
Morgantown, WV 26505
Tel. 304-285-5832
Fax. 304-285-6265

Original question:

Preconditioning is a procedure to load and unload the soft tissues
several times before the data collection. There were many great discussions
the preconditioning of soft tissue in YC Fung's early work (e.g.
Biorheology, 1973, vol 10, 139-155). The purpose to perform the
preconditioning, as I
understand, is to achieve more repeatable testing results. Although the
preconditioning has been routinely applied in soft tissue tests, I am
skeptical about the necessarity to perform such a procedure. If a soft
tissue is considered to be a viscoelastic material, and the purpose of
the testing is to determine the viscoelastic material properties, the
mechanical behaviour (e.g., stress relaxation behaviour) is varied
completely after
the preconditioning. In fact, the preconditioning behaviour can be predicted
using a viscoelastic model.

My question is:

1. Should preconditioning always be applied in any soft tissue tests:
tensile/compressional, stiffness, fatigue, stress relaxations tests ?

2. If the purpose of the preconditioning is to obtain a repeatable
result, can we do a large number of tests, instead of do

John Z. Wu, Ph.D.
National Institute for Occupational Safety and Health
Email: jwu@cdc.gov


This is a great question for Biomch-L.

I am not a biomaterials expert, but I tend to agree with you: if you have
a model that supposedly predicts history-dependent behavior, there
should be no need for preconditioning. In fact, data collected during
preconditioning could be extremely valuable if it is your goal to
estimate parameters for such models!

A.J. (Ton) van den Bogert, PhD
Department of Biomedical Engineering
Cleveland Clinic Foundation
9500 Euclid Avenue (ND-20)
Cleveland, OH 44195, USA
Phone/Fax: (216) 444-5566/9198

++++++++++++++++++++++++++++++++++++++++++++++++++ +

I have had the same questions as you regarding preconditioning of soft
tissues for testing. The basis for the procedure seems more empirical
than theoretical. Nevertheless, I have found it to be true that soft
tissue behavior becomes repeatable only after preconditioning. The
caveat is that tissue behavior is repeatable after a reasonable resting
period between tests, which is usually less than an hour in laboratory
testing. Perhaps tissue would display "unpreconditioned" behavior after
several days of rest, but it is not often feasible or desirable to wait
this long between tests. The commonly acknowledged limitation of QLV
theory is that information about short-time and long-time tissue
behavior is bounded by the time limits imposed during experimental
testing. The need to precondition tissue in order to obtain repeatable
results may simply reflect these experimental limitations. If you're
interested, I wrote a paper on ankle ligament viscoelasticity in which I
discussed these issues and attempted to roughly characterize the
behavior of the ligaments in the unpreconditioned state (Journal of
Biomechanical Engineering, 122(1): 15-22, 2000).

Good luck,

Jim Funk, Ph.D.
Biodynamic Research Corporation
9901 IH-10 West Suite 1000
San Antonio, TX 78230
Phone: 210-691-0281
Fax: 210-691-8823


The purpose of preconditioning (mechanical stabilization in some of the
spine literature) is to obtain repeatable testing. If you don't
precondition you will have a time dependance in your relaxation function,
G(t,t) because the first few cycles of loading will not have the same G(t)
(or other function depending on the test you perform). If you want to build
that into your analytic work, you can, but it will make your experimental
work cumbersome. Not proven, but frequently suggested is that awake active
people are preconditioned through most of their waking day.

Barry Barry S. Myers, M.D., Ph.D.
Associate Professor
Department of Biomedical Engineering
Duke University


************************************************** ******
Further Questions:

Dr. Funk,

Thank you for your response, and your good paper on viscoelastic
characteristics of ligaments. I agree with you that soft tissue should
be preconditioned to obtain a repeatable deformation behaviour.

However, I have a second question, how can a soft tissue be "properly"
preconditioned? According to YC Fung, the soft tissue should be
preconditioned at the same stress levels and loading states (uniaxial or
biaxial loading condition) as the subsequent testing in which data will
be collected. This obviously cannot be achieved in cases for loading at
large deformations, because the tissue will not only be preconditioned but
also be pre-damaged. Therefore, some researchers do the preconditioning at
20-30% of the largest strain levels of the final tests. Consequently, there
are no
repeatable test data if the preconditioning is performed at different
loading conditions. Is there a rule for the strain levels and loading
frequencies in preconditioning?


Yes, I've also heard that you should precondition to whatever strain
level and strain rate you plan on characterizing. I think this a good
idea. When I tested ankle ligaments, I preconditioned several specimens
to 20% strain at high rates without apparent ligament damage (no
increase in the toe region of the curve due to permanent deformation).
Of course, this leads to another thorny question about how to define
zero strain. I defined zero strain as the first point of measurable
resistance (2 N). Obviously, 20% strain from this point is not as much
stretch as if the zero point had been defined by, say, 50 N of

But to address your question more directly, I don't think that
preconditioning under the same loading conditions as subsequent testing
should present a problem. My opinion is that if you cannot precondition
a specimen to a certain strain level without damaging it, then you could
not viscoelastically characterize the tissue at that strain level,
anyway. The limit to how large a deformation you can characterize with
viscoelastic testing corresponds to the point at which measurable tissue
damage occurs after a small number of loading cycles. Viscoelastic
characterization requires a certain number of step loading cycles, and
if you're in the region where significant fatigue damage occurs after 5
cycles or creep damage occurs after 30 minutes, then your experimental
procedure will overly influence the tissue properties you're trying to
measure. In that case, you'd have to settle for a smaller deformation.
So I'm with Dr. Fung on this one, but that's just my two cents.

Good luck,
Jim Funk

************************************************** ******
Further Questions:


Thank you very much for your response. I agree with you that the
preconditioning behaviour should be separated from that of viscoelastic
behaviour, so that G(t) will reflect the characteristics of the material

However, in cases to test the soft tissue at large deformations, the
preconditioning cannot be performed at the same stress/deformation levels as
the subsequent testing, as suggested by YC Fung. Because the tissue will not
only be preconditioned but also be pre-damaged. Some researchers do the
preconditioning at 20-30% of the largest strain levels of the final tests,
this could result in un-repeatable test data. Is there a rule for the strain
levels and loading frequencies in preconditioning?

++++++++++++++++++++++++++++++++++++++++++++++++++ ++++++

I have never seen a rule or guidelines and have had the problem you describe
before. It occurs every time we do low load and failure testing. If you
find a solution, I would be very interested in it.


************************************************** ******

To unsubscribe send SIGNOFF BIOMCH-L to LISTSERV@nic.surfnet.nl
For information and archives: http://isb.ri.ccf.org/biomch-l