View Full Version : Summary of Replies: Tendon Gripping / Tensile Testing Techniques

Jeffrey Cartmell
12-14-1999, 09:21 AM
A few weeks ago, I posted a message seeking advice on soft tissue grips
for tensile testing of ligaments, specifically with regard to "freeze
clamping" methods. This message is a summary of the responses which I
received, as well as a repost of the original request. I thank everyone

who offered suggestions, your advice was most helpful.

- J. Cartmell


Original Message:

As part of my research project, I am planning to do mechanical testing
of rabbit patellar tendon specimens. I would like to do testing on the

tendon itself (without the bone ends attached), but may consider using a

bone-tendon complex. I am having difficulty achieving adequate gripping

of the samples on our Instron mechanical tester, and was wondering if
this group could be of help. A number of protocols utilizing custom
clamps and/or potting methods appear in the literature, specifically
utilizing liquid nitrogen.

I was wondering if anyone has information on:
1. What are the best methods for gripping the specimens if the bone
attachmenst are left intact ?
2. What are the best methods for gripping the specimens if the bone
attachments are removed ?
3. Are liquid nitrogen cooled grips for tissue esting comercially
available, and if so who is the supplier ?
4. Are there other methods of gripping tendon (either with or without
the bone attachments) which you can recommend?

I will be happy to list a summary of the responses to this query. Thank

you in advance for your help.



Debra Balderson (dsb13@york.ac.uk) wrote:

The best person to contact is Alan Goodship at the Royal Veterinary
College, North Mimms, Nr Hatfield, UK.

Richard Hall (r.m.hall@leeds.ac.uk) wrote:

I cannot offer any real good advice but the book by Mow and Hayes does
have some good pointers as does Nigg and Herzog - Anyway please inform
of your answers - I am interested in respect of another similar project

Saiwei Tung (swyang@bme.ym.edu.tw) wrote

Hi, for liquid nitrogen asked Dr. E.Y. Chao of U. Johns Hopkins,
Biomechanics Lab. , sine or cosine wave shaped gripper asked Dr. S.L.
Woo of U. Of Pittsburgh for help, or find from the ORS proceedings. you

can find contact phone numbers from their WEBsite

Cory Carter (carterc@SYNTHES.com) wrote:

If the bone plugs are attached, you can drill a hole through each plug
(maybe 1.5 mm, or perhaps 2.0 mm for multiple sutures) and pass
suture(s) through this hole. The suture can then be tied (in a
consistent manner) so that you now have a continuous loop of suture.
You can then pass a metal pin horizontally through this loop of suture
and secure the pin within the Instron machine.

If you want to remove the patellar and tubercle bone plugs, you can use
any of various standard suturing techniques (i.e., Mattress, Mason
Allen, etc.) through the ends of the soft-tissue. Then simply use the
same method described above to affix your suture-tendon construct to the


I believe Christian Gerber's article diagrams all of this quite nicely
(the Instron setup as well as various suturing techniques): "Mechanical

strength of repairs of the rotator cuff," in the Journal of Bone and
Joint Surgery, 1994, no. 75 or 76, I think.

Arik Sverdlik (bmesver@techunix.technion.ac.il) wrote:

I am just now finishing to write my M.Sc. thesis on tendon tensile
mechanical behavior, in which I did a lot of tensile tests on sheep
digital tendons. The issue of tendon clamping (without the bone/muscle
ends) is problematic and can be summed up in the sentence "A tendon is
not a wire". In my experiments I used flat sand-blasted metal clamps. I
add bellow the part about clamping from my "Methods" section, in which
you can find the reasons that led me to use such clamps.

Good Luck

Arik Sverdlik


The Clamps:

The clamps surface was flat. The surface of the fixed part of the clamp
was sandblasted, to a thin degree. The removable part had a water stable

thin sandpaper glued to it. The tow parts were adjoined by tow screws,
between which the tissue clamped length was placed.

Literature Review:

The issue of clamping tendons and ligaments in mechanical properties
experiments, was discussed in many sources and different solutions were
presented (Parker 1980, Ang 1989, Svendsen 1984, Jimenez 1989, Riemersa
1982, Liggins 1992). The main issues that are concerned are to preserve
the physiologic structure and condition of the tendon as much as
possible, in the vicinity of the clamping area (minimize clamps loads),
and to prevent slippage of the tissue from the clams (maximize clamps
loads). The main problem can be summed in the sentence "A tendons is not

a wire". Conventional used weirs can withstand longitudinal loads and
transversal loads of the same magnitude. Also, when clamping such weirs
to withstand longitudinal loads, the holding force is the friction
force, that appears due to transverse loading on the weir. Tendons are
different in tow ways. The first is that although they can withstand
high longitudinal loads, transverse loads cause squeeze of fluid and if
the load is not dispersed enough, it causes the disruption of the tissue

structure and damage to the tissue in the loaded area. Second, the
friction coefficients of tendons surface in their normal physiological
state is very low, about 0.07 (Moro-oka 1999), and therefor to hold with

a friction clamp a load on a tendon will require loads that are bigger
in an order of magnitude transversely on the tendon. In the beginning of

the current investigation, an attempt was done to clamp the tendons by
slipping the tendon through a fitting hole and tying a knot in the
tendon's end so on applying load, the knot will hold the tendon from
slipping. The possible advantage in knotting the tendon and holding on
the knot is that this action is supposed to lock the internal structure
of the tendon, in the clamping site. Results from loading to failure
experiments showed that the tendons were always torn in the knot site.
In other cases the loading made the knot advance on the tendon, living
behind a disrupted squeezed tendon length that finally caused the tare
of the tendon. Also, knotting the tendon caused deformation of the
tendon to about 5 diameter lengths from the knot, which implies to a
probable disruption in the internal structure of the tendon. Therefor
the knotting technique was abandoned.
All of the clamping solution causes the local change in tendon
properties, in the clamped area, so the tendon can withstand the
transferred transverse loads. Some use cryo clamps that cause a local
freezing of the tendon in the clamps and makes it much harder so it can
withstand the clamping loads (Riemersa 1982, Liggins 1992). This method
can not be used in humidity and temperature control chambers and
therefor was not applicable in out case. Most of the reported
investigations that used humidity and temperature control chambers used
different sorts of flat clamps, or teethed clamps (Pradas 1990, Wang
1991, Wang 1995). In this kind of clamps, the clamped area goes
irreversible damage and the liquid is squeezed out and the tissue
section in the clamps is actually dryad out. The latter causes the
clamped area to be harder and more resistant to transverse loading. If
the clamped area is large enough, it allows the applied load to be
dispersed on a large area and thus prevent
the tare of the tissue in the clamped region. Others used tow concentric

clamps to allow spreading the tendon on a large area (Svendsen 1984,
Jimenez 1989).
After clamping a tendon in the clamps, the section of the tendon that
was closest to the clamp as 5 diameter sections, was twisted up to
angels of 45* from the initial longitudinal line of the tendon.
Releasing the clamp revealed that the clamped area was flattened to a
sheath like area with a length of about 6 times the tendons initial
section diameter. This sheath like area was dry, hard, and
half-transparent, with a yellowish color, resembling air dryad tendons.
The clamping procedure started with an initial clamping. The screws were

tightened and then released and the clamps were opened. The flattened
surface was separated from the clamp surface and realigned and centered.

Finally the clams were tightened again to the maximal possible
tightening force. There was no case in which the clamps tore the tendon,

by the tightening. It should be noted that on loading, no slippage was
apparent in the ands of the experiments, the clamped region remained
clear and intact.
It is important to note that the tendon specimens had sometimes a region

of more cartilage-like behavior, apparent in resistance to bend. Care
was taken that these sections of the tendons will not be presented in
the free length between the clams, but only in the clamping region

Chuck Pell (cap@ntrobotics.com) wrote:

Use knots --- with a pin through the top of the knot
distal to the region of interest. Tie the knots on the far
side of a standard hole grip.

Robert Day (robday@rph.health.wa.gov.au) wrote:

We have done some testing on both bone/tendon/bone preparations and free

tendons in tension. A free tendon is difficult to hold, but if you have

enough length it can be done. We manufactured small wedge action grips
with interdigitating involute teeth to hold the tendon. Try to avoid
sharp edges and hard corners. The total grip length was about 20 mm,
and the free length about 60 mm (the tissue was from dogs). We got most

of our failures in the mid substance of the tendon, but there was a
significant initial slip within the grips, which then settled down. An
extensometer across the gauge length gets around this problem.

For bone blocks, we usually embed them in PMMA and then drive a small
wire or pin through the PMMA/bone block. For bigger blocks we use PMMA
to embed in a metal container (square tube works well), and then pin the

lot before fixing the metal container to the grips.

We have not yet tried ice vice type systems, but we would like to in
future when we get the time ;-). All our equipment was made up in our
lab here.

Ashvin Thambyah (ashvin@nus.edu.sg) wrote:

we've been quite successful in using sandpaper (as an interface to the
machine clamps) to grip monkey patella tendon for testing to failure. A
majority of our tests resulted in midsubstance failure with
load-deformation graphs showing that no slippage had taken place.

Eric Powel (epowell@fs1.ho.man.ac.uk) wrote:

For our work we used liquid nitrogen cooled clamps. We had to have them

milled from solid copper by our workshop. Some details are contained in

our paper:

Powell, ES et al (1989) Non-suture repair of tendons. J Biomed Eng
11 (3), 215-218

Apart from the gripping surfaces the rest was covered with polyurethane
foam to slow down the rate at which the grips warmed up. If you want
any further details please contact me.

John DesJardins (jdesjar@clemson.edu) wrote:

At the Johns Hopkins University Biomechanics Lab, we have had the
opportunity to develop a novel Cryo-clamp design that has worked very
well for us and is relatively inexpensive as compared to liquid Nitrogen


The design consists of machining up two thin aluminium plates (1/4 to
1/8 inch thick) and putting groves on the inside faces. We used V shaped

groves that fit into one another with a depth of about 1/8 inch. Not too

sharp or you can cut the ligament. Then using 4 bolts tighten the
ligament between the plates only ever so slightly. This is not what
holds the ligament, so don't tighten it very much at all. Then, take a
small block of dry ice, and place it one one of the outside faces of the

clamp. We designed some small spring loaded plastic containers that both

held the ice in place, insulated it, and pressed it against the face of
the clamp. That'a all there is to it. We had one end of the clamp with a

hinge joint so that we could attach it to theactuator, but any interface

will do. The ice block (1inch x 0.5 inch x inch lasted about 15minutes
and then we just put another in. The frozen clamp could hold more than
5000N, but we never did test it to failure. Once the clamp unfroze and
the ligament just slipped right out, but that was our fault. The
freezing never had any bad effects on the ligament and we were always
very happy with the design. You could use muscle as well as ligament in
the clamp, and aside from the small amount of extra weight that the
clamp added to the system, it was perfect. Try it out, I highly
recommend using dry ice. It cost us about 50 cents a pound and we
usually got 10 pounds for a complete day of testing. We cut it into
blocks using a regular band saw, but watch out that you give the band
saw a rest every minute or so, of you will freeze the blade and it will

We used these clamps for a number of experiments to hold the quadriceps
during physiologic loading experiments and this work is published in
the following references:

J.D. Des Jardins, B.A. MacWilliams, D.R. Wilson, E.Y.S. Chao, The
In-VitroAssessment of Knee Joint Kinematics Under Physiologic Loading.
Proceedings of the 43rd Annual Meeting of the ORS, February 9-13, 1997,
San Francisco, California.

B.A. MacWilliams, J.D. Des Jardins, D.R. Wilson, J. Romero, E.Y.S. Chao,

A Repeatable Alignment Method and Local Coordinate Description For Knee
Joint Testing and Kinematic Measurement. Journal of Biomechanics, Volume

31, Number
10, pp. 946-950, 1998.

J.J. Elias, J.D. DesJardins, A.F. Faust, S.A. Lietman, E.Y.S. Chao, Size

and Position of a Single Condyle Allograft Influence Knee Kinematics.
Journal of Orthopaedic Research, Vol. 17, No. 4, 1999.

B.A. MacWilliams, D.R. Wilson, J.D. DesJardins, J. Romero, E.Y.S. Chao,
Hamstrings Cocontraction Affects Knee Kinematics and Kinetics during
Weightbearing Flexion. Journal of Orthopaedic Research, Accepted for
Publication, 1999.

A.M. Bevilacqua, J.D. Des Jardins, D.R. Heekin, A.F. Faust, J.J. Elias,
E.Y.S. Chao, The Effect of Tibial Tray Malrotation on Knee Kinemaitcs in

Total Knee Arthroplasty. Journal of Arthroplasty, Submitted for
Publication, 1998.

D.R. Wilson, B.A. MacWilliams, J.D. Des Jardins, E.Y.S. Chao, In-Vitro
Simulation of Knee Joint Kinematics. Proceedings of the 1996 ASME
Mechanics and Materials Conference, June 12-14, 1996, Baltimore,

D.R. Wilson, B.A. MacWilliams, J.D. Des Jardins, E.Y.S. Chao, Effects of

Hamstrings CoContraction on Tibial Position. Proceedings of the 43rd
Annual Meeting of the ORS, February 9-13, 1997, San Francisco,

J.D. Des Jardins, J.J. Elias, A.F. Fuast, P. Shih, F.J. Frassica, E.Y.S.

Chao, Influence of Allograft Size and Position on Knee Kinematics.
Proceedings of the Annual Meeting of the International Society on Limb
Salvage (ISOLS), September 10-12, 1997, New York, New York.

A.M. Bevilacqua, J.D. Des Jardins, D.R. Heekin, A.F. Faust, J.J. Elias,
S. Yang, P.Rullkoetter, E.Y.S. Chao, The Effect of Tibial Tray
Malrotation on Knee Kinemaitcs in Total Knee Arthroplasty Proceedings of

the 10th Annual Symposium of the International Society for Technology in

Arthoplasty (ISTA), September 25-27, 1997, San Diego, California.

A.M. Bevilacqua, J.D. Des Jardins, D.R. Heekin, A.F. Faust, J.J. Elias,
S. Yang, P. Rullkoetter, E.Y.S. Chao, The Effect of Tibial Tray
Malrotation on Knee Kinemaitcs in Total Knee Arthroplasty. Proceedings
of the Meeting of the Society of Military Orthopaedic Surgeons (SOMOS),
October 6-10, 1997, Lake Placid, New York.

A.M. Bevilacqua, J.D. Des Jardins, D.R. Heekin, A.F. Faust, J.J. Elias,
P. Rullkoetter, E.Y.S. Chao, The Effect of Tibial Tray Malrotation on
Knee Kinemaitcs in Total Knee Arthroplasty. Proceedings of the 31st
Annual Residents Conference of the American Orthopaedic Association
(AOA), February 19-22, 1998, Sacramento, California.

J.D. Des Jardins, A.F. Faust, J.J. Elias, S. Lietman, E.Y.S. Chao, The
Influence of Segmental Femoral Allograft Size and Position on Knee
Kinematics. Proceedings of the 44rd Annual Meeting of the ORS, March
16-19, 1998, New Orleans,

J.D. Des Jardins, B.A. MacWilliams, D.R. Wilson, S. Kato, J.J. Elias,
E.Y.S. Chao, The Kinematic Effects of Hamstring Cocontraction on Intact
and ACL Deficient Knees. Proceedings of the 44rd Annual Meeting of the
ORS, March 16-19, 1998, New Orleans, Lousiana.

A.M. Bevilacqua, J.D. Des Jardins, D.R. Heekin, A.F. Faust, J.J. Elias,
E.Y.S. Chao, The Effect of Tibial Tray Malrotation on Knee Kinemaitcs in

Total Knee Arthroplasty. Proceedings of the 44rd Annual Meeting of the
ORS, March 16-19, 1998, New Orleans, Lousiana.

Mark Zobitz (Zobitz.Mzrk@mayo.edu) wrote:

A design of a "cryo-jaw" that we use is described in:
Riemersa DJ. Schamhardt HC.
The cryo-jaw, a clamp designed for in vitro rheology studies of horse
digital flexor tendons.
Journal of Biomechanics. 15(8):619-20, 1982.

We use liquid CO2 to freeze the tendon or muscle end. I think one tank
costs around $15 and lasts quite awhile. We have used this device
successfully to grip tendon and muscle at least to 500lbs. One drawback

is that care must be used so that only the clamped portion of the tendon

is frozen.

If a bone is attached, PMMA seems to work the best for me for potting.
The addition of a k-wire through the bone really helps the fixation.

Melissa Brown (melissa_brown@neucoll.com) wrote:

Perhaps you can contact John West. My latest address for him is at the
University of Utah (John.West@hsc.utah.edu). He ran some testing on
rabbit tendons for me a few years back using freeze clampswhen he was at

Orthopedic BIomechanics Institute in Salt Lake City. This was a custom
set-up that he loaded with dry ice just prior to testing. They worked
well but it could be
tricky getting an exact length between the grips. Good luck.

Eric Martz (martz@osteotech.com) wrote:

I have tested some other types of animal tendons, and found that they
are indeed very difficult to grip. The fixtures that I ultimately used
had saw-tooth pattern to really grab the tendons. I did not try
freezing the grips (not practical at the time, and I was doing
quick-and-dirty testing). I think that you'll reach a point in the test

that the tendon will simply slip out of the grips, and I remember seeing

some results in the literature that also describe that type of slippage.


Thank you once again to all who helped. I hope this summary will be of
use to others.
- J. Cartmell

Jeffrey S. Cartmell
Ph.D. Candidate
Biomedical Engineering
Rutgers University and UMDNJ
email: cartmejs@umdnj.edu
phone: (732) 235-6954
fax: (732) 235-6002

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