Scand J Med Sci Sports 1997: 7: 6 2 4 6 Copyright 0 Munksguurd 1997

Printed in Denmark . AN rights reserved Scandinavian Journal of

M E D I C I N E & SCIENCE
IN SPORTS
ISSN 0905-7188

Function and biomechanics of tendons

Kirkendall DT, Garrett WE. Function and biomechanics of tendons. D. T. Kirkendall’, W. E. Garrett2
Scand J Med Sci Sports 1997: 7: 62-66. 0 Munksgaard, 1997 ’Department of Physical and Occupational
Therapy, *Division of Orthopedic Surgery,
Tendon is a highly organized connective tissue joining muscle to bone, Duke University Medical Center Durham,
capable of resisting high tensile forces while transmitting forces from North Carolina, USA
muscle to bone. The dense, regularly arranged collagenous tissue is made
up of fibers, cells of various shapes and ground substance. The mechanical
and physiological characteristics of collagen (nearly 85% of the dry weight
of tendon) dictate the qualities of tendon. In addition, tendon is flexible
so that it can bend at joints, as well as acting as a damping tissue to
absorb shock and limit potential damage to muscle (1). Tendon also shows
a degree of extensibility. If the strain used to stretch a tendon could be
recovered, a beneficial elastic effect would be achieved. Muscles lengthen
Key words: tendon; chemistry; mechanics;
and shorten in a cyclical manner. During the lengthening period, elastic plasticity
energy can be stored and used as elastic recoil. For example, the Achilles
tendon is stretched late in the stance phase as the triceps surae muscles Donald T. Kirkendall, Box 3435, Duke University
contract and the ankle dorsiflexes. Prior to plantarflexion, muscle acti- Medical Center, Durham, NC 27511, USA
vation ceases and stored energy helps to initiate planter flexion. Accepted for publication 2 December 1996

vascular tendons are encased in a paratenon and its

Basic structure
Tendon (Fig. 1) is made up of densely packed fibers
of collagen that extend over its entire length and are
arranged in parallel to its long axis. The collagen
fibers comprise thinner fibrils and the few fibroblasts
are centrally located. Groups of fibers are arranged
in a fascicle and each tendon is then made up of
multiple fascicles.
Surrounding the tendon is a membrane called the
epitenon that is not unlike a synovium. The inner sur-
face of the epitenon is continuous with the endo-
tenon, which binds the collagen and contains neural,
vascular and lymphatic supply. In some tendons, a
loose areolar tissue, the paratenon, surrounds the epi-
tenon. The presence of elastic fibers in the paratenon
allows for stretch especially in tendons that move
along a straight path. The paratenon can be replaced
by either a synovial sheath or a bursa and is some-
times referred to as the tenosynovium, as found in
the tendons of the flexor muscles of the wrist and
hand. Without the described sheath, the paratenon
may be termed the tenovagium. The epitenon and pa-
ratenon make up the so-called peritendon.
Tendon is fairly well vascularized; more so than
cartilage, less so than muscle. Blood arrives to the
tendon through connections at the perimysium, peri-
osteum and other surrounding tissues. These vessels
then pass through the paratenon. Tendons may be
characterized as being ‘vascular’ or ‘avascular’. The
accompanying vessels. Avascular tendons are envel-
oped in a sheath, which functions as a conduit lead-
ing to the need of diffusion through synovial fluid to
provide nutrients to tendons.
The nervous innervation of tendon is confined to a
sensory role offering proprioceptive input from me-
chanoreceptors found near the muscle-tendon junc-
tion. Tendon has no motor function so it receives no
efferent innervation.
The connection of the tendon with the muscle or
bone is a result of the endotenon being continuous
with the perimysium and periosteum.
At the insertion to bone (the osteotendinous junc-
tion), the collagen fibers enter bone as Sharpey’s
fibers. The simple attachment to bone occurs when
fibrils pass into bone directly into fibrocartilage with
little contact with the periosteum. A more complex
method involves superficial fibrils inserting into the
periosteum and deeper fibrils inserting directly into
bone.
Chemistry
The main fibroblast responsible for the production
of tendon collagen and its matrix is the tenocyte, a
cylindrical shaped cell capable of producing matrix
precursors, elastin, proteoglycans and collagen.
While there are few of these tenocytes, there is ample
ground substance and some elastin.

62
 Function and biomechanics
64nm
Collagen
More than 12 different types of collagen have been
recognized, but it is easy to think of collagen as either
fiber forming or non-fiber forming. Types I, I1 and
I11 are fiber-forming collagens and make up most of
i$ Hole Zone
! I '
the tendon collagen secreted into extracellular spaces,
which eventually forms fibrils. Of these, type I makes
Packing of
MoleCuleS
-
-I-
-
-;
up about 90% of the collagen in the body and is the
primary collagen of tendons. The other collagens are
non-fiber-forming collagens. Articular cartilage has
Collagen 280nrn
primarily type I1 collagen. Types IV and V make up Molecule '
,-I

I
basement membrane collagens.
The tenocyte produces a precursor molecule, pro-
collagen, that is secreted, then cleaved to form tropo-
collagen. Collagen fibrils are then assembled via non- a2
covalent cross-links. Covalent cross-links then bind
fibrils into the typical triple helix yielding the body's
strongest protein (Fig. 2).
Triple a,
Helix
at 1
1.5nm

Elastin
Elastin is required in structures that undergo great
changes in length in the absence of any permanent
change in structure. There is only a small amount of
elastin in the tendons of the extremities and much
more in places like the ligamentum nuchae. In ten-
dons, elastin is less than 1% of dry weight (versus the
aorta where elastin makes up 30-60% of dry weight).
The unique properties of elastin are due to the link-
age of unique, copper-dependent, amino acids, de-
smosine and isodesomosine, to lysine residues.
Ground substance
Much of the viscoelastic properties of tendon are due
to its ground substance. The water-binding capacity
of proteoglycans, glycosaminoglycans, plasma pro-
teins, and other small molecules help stabilize the col-
lagen skeleton of tendon. Those portions of the ten-
don that experience the greatest tensile forces have
low proteoglycan content and high rates of collagen
.~
Paratenon
Ep,tenon
Endofenon
~Fibroblesl
. . ~i
~~
a Collagen a-chain (purity 10 molecules)
Structure of Collagen
Fig. 2. The microstrucure of collagen. Three alpha chains coil
to form a triple-stranded helical rod (from ref. 1).
synthesis as opposed to areas that undergo frictional
and compressive forces (2).
A secondary component of tendons is proteo-
glycan, which comprises 1-209'0 of the dry weight of
tendon, depending on factors such as age, site, and
the history of mechanical loading. Proteoglycans rep-
resent a diverse family of glycosylated proteins that
contain numerous sulfated polysaccharides (glycosa-
minoglycans, or GAGs). The predominant proteo-
glycan component in tendon is a low molecular
weight dermatan sulfate. Larger molecular weight
proteoglycans are present in the fibrocartilaginous re-
gions of the tendon, which are placed under compres-
sive loads in vivo.
The physical properties of proteoglycans are dic-
tated by the presence of a large number of negatively
charged GAGs, which attract counter-ions and water
molecules in the tissue. These characteristics are be-
lieved to contribute to the tissue's compressive prop-
,
i Peritendon erties and viscoelastic behavior.

~~
Fiber bundle
Fiber Tendon biomechanics
Fibnl

~ Microfibril
Tendons must be able to transmit high muscle forces
to the skeleton for movement, and the parallel ar-
Fzg I Structural model of a tendon. rangement of fiber bundles is quite efficient at this

63
Kirkendall & Garrett
Deformation ( ative tendinopathy, mucoid degeneration, tendolipo-
Itime
Deformation
I in tendon length. Conversely, control of fine motor
stress
Stress relaxation
time
Fig. 3. Graphic representation of the biomechanical properties
of tendon demonstrating the concepts of creep and stress-relax-
ation (from ref. 1).
function. However, tendons have a low resistance to
shear forces. Thus, tendons are designed to transmit
forces with minimal deformation or energy loss.
Collagenous tissues such as ligament and skin
show a similar stress-strain relationship to that of
tendon. Two primary features are stress-relaxation
(decreased stress over time with constant deforma-
tion) and creep (increased length over time with a
constant load) (Fig. 3). These properties lead to a
load-deformation relationship for tendon that is de-
pendent on the activities prior to assessment (history-
dependent properties).
A typical load-deformation curve is presented in
Fig. 4. Four distinct features of the curve can be dis-
cussed in light of anatomic microstructure (3). The
initial ‘toe’ region corresponds to the arrangement of
the fibers with the direction of the stress,
straightening out the initial ‘waviness’ of the fiber
bundles. The linear portion of the curve (the elastic
stiffness of the tendon) is a result of the elongation of
the helical structure. At the end of the linear portion,
unpredictable failure begins, leading to eventual rup-
ture and a recoil of tendon at the ruptured end. In
vivo, there is some margin for error as the maximum
isometric force of a muscle is about one-third that
necessary for tendon failure. However, repetitive sub-
maximal loading and fatigue can lead to tendon
failure.
It should be noted that normal tendon does not
fail in response to strain. When this muscle-tendon-
64
bone unit fails, the typical locations are bone, bone-
tendon junction or muscle-tendon junction. How-
ever, diseased tendon can fail. Kannus & Jozsa (4)
showed that 97% of nearly 900 patients with spon-
taneous ruptures of tendon exhibited degenerative
changes in the ruptured tendon (e.g. hypoxic degener-
matosis or calcifying tendinopathy).
Tendons function to transfer tension from muscle
to bone. According to Cutts et al. ( 5 ) , when the
muscle contracts, the resulting stretch by the tendon
leads to three definite outcomes.
First, compliance of a tendon makes it difficult to
hold a joint steady as changes in force lead to changes
tasks is easier because the force necessary for change
is reduced from small changes in muscle length.
Second, stretched tendons store elastic energy that
is released upon recoil effectively reducing the work
of the muscle.
Third, stretching of the tendon requires that the
muscle may have to shorten more than might nor-
mally be necessary. Therefore, this extensibility can
be considered advantageous, such as when the tendon
acts like a spring, or disadvantageous, as when energy
is being transfered to an external system.
Ker et al. (6) argue that tendon function should
be considered in concert with muscle function and
describe a combined mass of the muscle and tendon.
Thin tendons (those which stretch the most) require
longer muscle fibers to yield the ideal ratio of muscle
mass to tendon mass. Their optimum ratio of 34
yields a stress of 10 MPa, which is well below the 100
MPa breaking point for normal tendon, demonstrat-
ing that mammalian tendons are far thicker than
necessary.
Ker et al. (6) also point out that the safety factor
ELONGATION
Fig. 4 . A typical load-elongation curve demonstrating the ‘toe’
or primary region (l), the ‘linear’ or secondary region (2) and
the end of the secondary region (3) (from ref. 1).
(the difference between tendon stress and breaking
point) is reduced in muscles that supply elastic recoil
during locomotion. For example, they report that the
stress on the Achilles tendon is as high as 67 MPa,
demonstrating that the safety factor for this tendon
is low in comparison with other tendons. Adding de-
generative changes to the tendon will effectively raise
the muscle-to-tendon mass ratio and the breaking
force of tendon will approach the stress forces, lead-
ing to rupture.
Plasticity of tendon
A variety of conditions may alter the normal charac-
teristics of tendon. These might include immobiliza-
tion, exercise, aging and medications.
Immobilization
Like ligament, immobilization results in profound re-
ductions in the mechanical properties of tendon.
Most noticeably, there is a decrease in tendon
strength along with an increase in collagen turnover
(7, 8). Like so many other tissues, tendon is inti-
mately affected by immobilization.
Exercise
The literature on the effects of exercise training on
tendon function is indecisive. Some studies have
shown increases in tendon strength (9-1 1) while a
long-term study (1 year) of training in pigs showed no
change in the mechanical properties, area or collagen
content of flexor tendons (12), while improvements in
strength, area and collagen content were demon-
strated in extensor tendons (1 1). Further, while an
increase in load to failure was found in the tendons
of trained animals, no differences were seen for
weight, water or collagen content (10). Structurally,
there appears to be an increase in collagen synthesis,
as smaller fibril bundles have been documented (13).
Woo et al. (14) proposed a model of the mechanical
responses of tendon to immobilization and exercise
(Fig. 5). They suggest that tendon undergoes pro-
found insult as a result of immobilization, but train-
ing has little or no effect on these mechanical prop-
erties.
Aging
Tensile strength reaches a plateau after collagen
maturation and declines with age. This decline is re-
lated to a decrease in both insoluble and total colla-
gen content (15). The loss of collagen and its cross-
linking leads to an increase in tendon stiffness (16,
17). A variety of degenerative conditions can affect
the function of tendon. Of interest is the fact that
t
I
- Decrease
Immobilization
tress
I
Increase
Stress
Physiologic
Activities I
-
Function and biomechanics
Exercise
v)
w
t
p:
STRESS AND STRAIN
DURATION
Fig. 5. The hypothetical relationship of the effects of stress and
motion on the responses of soft connective tissues (from ref.
14).
spontaneous rupture of tendons does not occur in the
absence of some pre-existing histopathological con-
ditions (4). In nearly 900 patients, degenerative
changes were seen in practically all those with spon-
taneous rupture (4). The most common conditions
seen were hypoxic degenerative tendinopathy (44%),
mucoid degeneration (21%), tendolipomatosis (8%)
and calcifying tendinopathy (5%).
Medication
Corticosteroids. Glucocorticoids are frequently used
in the treatment of sports injuries due to their potent
anti-inflammatory effects. Glucocorticoids inhibit the
synthesis of collagen (18) and occasionally local injec-
tions around patellar and Achilles tendons may pre-
cede a tendon rupture (19, 20). The systemic effects
of oral glucocorticoid treatment have been demon-
strated in patients with bilateral Achilles tendon rup-
ture (21, 22).
Laboratory studies are less conclusive due to the
common variables of dosage, duration of therapy,
location of injection, or type of drug. For example,
local injection of hydrocortisone acetate (20 mg/kg)
every third day for 24 days showed increases in pero-
neal tendon strength in the absence of changes in col-
lagen content, while the same injection schedule into
knee joints lead to reduced tensile strength of the in-
terface between the posterior cruciate ligament and
bone (23). Daily injections of prednisolone (2 mg/kg)
near the peroneal tendons for 2 weeks increased the
maximum load; stress and energy absorption were in-
65
Kirkendall & Garrett
creased as was elastic stiffness, the latter due to more
stable cross-linking within the tendon (24). Longer
term injection schedules (10 mg/kg cortisone every
third day for 55 days) did not alter mechanical prop-
erties, but the hydroxyproline content was reduced
(25).
It is likely that corticosteroids impact tendinous
tissue in two ways. During the first week or two, corti-
costeroid injection induces a rapid increase in the ten-
don’s mechanical stability due to a change in the
cross-linking pattern. Then, an inhibition of protein
synthesis leads to collagen thinning, to a progressive
reduction in collagen content (1, 25).
Non-steroidal anti-inflammatovy agents. It is well
known that indomethecin increases the tensile
strength of tendons and total collagen content (26).
Similar results were found in tendons undergoing re-
pair (27), suggesting that much of the increase in ten-
don strength is due to improved collagen cross-
linking (1).
Acknowledgement
The authors wish to acknowledge the contribution of Dr Far-
shid Guilak in the Orthopedic Research Laboratoy of the Divi-
sion of Orthopedics at Duke University Medical Center for his
assistance in the preparation of this manuscript.
References
1. Best TM, Garrett Jr WE. Muscle and tendon. In: Delee
JC, Drez D, ed. Orthopaedic sports medicine: principles
and practice. Philadelphia, PA: WB Saunders, 1994: 1-59.
2. Banes AJ, Link GW, Bevin AG et al. Tendon synovial cells
secrete fibronectin in vivo and in vitro. J Orthop Res 1988:
6: 73-82.
3. Abrahams M. Mechanical behavior of tendon in vitro: a
preliminary report. Med Biol Engr 1967: 5: 3 3 4 3 .
4. Kannus P, Jozsa L. Histopathological changes preceding
spontaneous rupture of a tendon. J Bone Joint Surg 1991:
73-A: 1507-25.
5. Cutts A, Alexander RM, Ker RF. Ratios of cross-sectional
areas of muscles and their tendons in a healthy human
forearm. J Anat 1991: 176: 13347.
6. Ker RF, Alexander RM, Bennett MB. Why are mam-
malian tendons so thick? J Zoo1 1988: 216: 309-24.
7. Amiel D, Woo SL-Y, Harwood FL, Akeson WH. The ef-
fect of immobilization on collagen turnover in connective
tissue: a biochemical-biomechanical correlation. Acta Or-
thop Scand 1982: 53: 325-32.
8. Noyes FR. Functional properties of knee ligaments and
alterations induced by immobilization. Clin Orthop 1977:
123: 21042.
66
9. Tipton CM, Schild RJ, Tomanek RJ. Influence of physical
activity on the strength of knee ,ligaments in rats. Am J
Physiol 1967: 212: 783-7.
10. Viidik A. The effects of training on the tensile strength of
isolated rabbit tendons. Scand J Plas Reconstr Surg 1967:
1: 14147.
11. Woo SL-Y, Ritter MA, Amiel D et al. The biomechanical
and biochemical properties of swine tendons. Long-term
effects of exercise on the digital extensors. Connect Tissue
Res 1980: 7: 177-83.
12. Woo SL-Y, Gomez MA, Amiel D et al. The effects of exer-
cise on the biomechanical and biochemical properties of
swine digital flexor tendons. J Biomech Eng 1981: 103: 51-
6.
3. Michna H. Morphometric analysis of loading-induced
changes in collagen-fibril populations in young tendons.
Cell Tissue Res 1984: 236: 465-70.
4. Woo SL-Y, Gomez MA, Woo Y-K, Akeson WH. Mechan-
ical properties of tendons and ligaments. 11. The relation-
ships of immobilization and exercise on tissue remodeling.
Biorheology 1982: 19: 397408.
5. Vogel HC. Influence of maturation and age on mechanical
and biochemical parameters of connective tissue of various
organs in the rat. Connect Tissue Res 1978: 6: 161-6.
16. Bailey AJ, Robins SP, Balian G. Biological significance of
the intermolecular crosslinks of collagen. Nature 1984:
251: 105-9.
17. Eyre DR. Crosslinking in collagen and elastin. Ann Rev
Biochem 1984: 53: 71748.
18. Nimni ME, Bavetta LA. Collagen synthesis and turnover
in the growing rat under the influence of methylprednis-
olone. Proc SOCExp Biol Med 1964: 117: 618-23.
19. Lee HB. Avulsion and rupture of the tenocalcaneus after
injection of hydrocortisone. Br Med J 1957: 2: 395.
20. Oxlund H. Changes in connective tissues during cortico-
tropin and corticosteroid treatment. Thesis. Aarchus: Insti-
tute of Anatomy, University of Aarchus, 1983.
21. Lee MLH. Bilateral rupture of Achilles tendon. Br Med J
1961: 1: 1829-30.
22. Melmed EP. Spontaneous bilateral rupture of the calcaneal
tendon during steroid therapy. J Bone Joint Surg 1965:
47B: 1 0 4 5 .
23. Oxlund H. The influence of a local injection of cortisol on
the mechanical properties of tendons and ligaments and
the indirect effect on skin. Acta Orthop Scand 1980: 51:
231-8.
24. Oxlund H, Manthorpe R, Viidik A. The biomechanical
properties of connective tissue in rabbits as influenced by
short-term glucocorticoid treatment. J Biomech 1981: 14:
129-33.
25. Oxlund H. Long-term local cortisol treatment of tendons
and the indirect effect of skin. Scand J Plast Reconstr Surg
1982: 16: 61-6.
26. Vogel HC. Mechanical and chemical properties of various
connective tissue organs in rats as influenced by nonster-
oidal antirheumatic drugs. Connect Tissue Res 1977: 5: 91-
5.
27. Carlstedt CA, Madsen K, Wredmark T. The influence of
indomethecin on tendon healing. A biomechanical and
biochemical study. Arch Orthop Trauma Surg 1986: 105:
332-6.