REVIEW ARTICLE
The Basics of Soft Tissue Healing and General Factors
that Influence Such Healing
Kevin A. Hildebrand, MD, Corrie L. Gallant-Behm, BSc, Alison S. Kydd, BSc, and David A. Hart, PhD
Abstract: Wound healing and repair of injured tissues follows
several steps in the healthy individual. Following birth, the process is
initiated by the inflammatory response and subsequent steps are
based on this initial response. Whereas wound healing generally leads
to a repair of the injured site, it does not lead to tissue regeneration.
This difference between repair and regeneration has influence on tis-
sues such as ligaments and tendons that function in a mechanically
active environment. Thus, the dynamic interface between mechanics
and biology influence the effectiveness of the healing response. The
biology of the host is also influenced by a variety of factors including
age, gender, genetics, and tissue history, factors that impact the
outcome of the healing response. This review focuses on several of
the issues surrounding the healing of tissues in general and ligaments
and tendons more specifically. In addition, the use of interventions
such as cell and gene therapy to improve healing is discussed.
Key Words: tissue repair, tissue regeneration, phases of healing,
genetic manipulation, gender and hormones, improved healing, tissue
heterogeneity in healing, response of healing to environment
(Sports Med Arthrosc Rev 2005;13:136–144)
GENERAL ASPECTS OF THE BASIC SCIENCE OF
TISSUE HEALING AND REPAIR
Wound healing following overt injury to a tissue in the
skeletally mature individual follows some general ‘‘rules’’
irrespective of the tissue involved. The details of the healing
response may be influenced by several factors including the
tissue source (ie, skin, lung, joint location-intra versus extra
articular, etc), extent of functional loss, health of the individual
(ie, nutrition), co-morbidities (ie, diabetes), as well as some
others (female vs. male), some of which will be discussed later
in this chapter. In general, similar steps occur that lead to the
‘‘orderly’’ repair of the tissue, and in many cases allows
a return to at least partial function. However, in some cases, the
repair process can lead to a loss of function (ie, scarring of the
heart, muscle, lung, kidney, and liver). Over the past several
years, fairly intense investigation has been focused on the healing
From the McCaig Centre for Joint Injury and Arthritis Research, Alberta Bone
and Joint Health Institute, University of Calgary, Alberta, Canada.
Reprints: Dr. Kevin A. Hildebrand, MD, Division of Orthopaedics, Department
of Surgery, McCaig Centre for Joint Injury and Arthritis Research, Alberta
Bone and Joint Institute, Faculty of Medicine, University of Calgary, 3330
Hospital Drive, NW Calgary, Alberta, Canada T2N 4N1 (e-mail:
hildebrk@ucalgary.ca).
Copyright Ó 2005 by Lippincott Williams & Wilkins
136 Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
response of many connective tissues that function in a set of
mechanically diverse environments. These include ligaments,
tendons, menisci, and joint capsules. Based on the outcomes of
these studies, it is clear that the functional outcome of the
healing process in such tissues appears to follow the ‘‘general’’
rules, but there are certainly some unique tissue-specific fea-
tures that contribute to the outcomes. The most obvious out-
come is ‘‘repair’’ rather than ‘‘regeneration’’, and the extent of
repair depends in part on many variables including those men-
tioned previously.
The basic, and somewhat arbitrarily defined steps in the
repair or healing of a tissue following overt injury are: (1)
hemostasis and a rapid inflammatory phase; (2) a phase of cell
proliferation and matrix deposition; and (3) a slow remodeling
phase, which may take up to months or years.1–3
Studies in animal models of MCL injury4–6 and porcine
models of skin healing7–9 have revealed a similar progression
of steps in the repair process at the molecular and histologic
level, but with some tissue-specific details.
The Inflammatory Phase
Acutely following overt injury in an otherwise healthy
adult, there is usually pain and bleeding into the area of injury
that may be variable depending on the site of injury. He-
mostasis is restored by the formation of a fibrin clot, which
prevents further bleeding and serves as a provisional matrix for
migrating cells. Blood clotting also results in the release of
pro-inflammatory molecules including components of the
clotting cascade and cytokines and growth factors released
from cells such as platelets.1 Some of these released factors are
chemotactic for other cells such as polymorphonuclear leuko-
cytes and monocytes, which migrate into the wound site and
further amplify the cascade. This influx of additional cells also
contributes to the subsequent migration, localization, pro-
liferation, and activation/differentiation of other cells (circu-
lating or tissue mesenchymal stem cells, fibroblast-like cells,
and others), which sets the stage for the second phase of the
repair process.
The Matrix Deposition Phase
Following consolidation of the fibrin clot and the
temporally regulated influx, proliferation and activation of
a subset of cells, deposition of matrix molecules designed to
bridge the damaged area with the residual endogenous lig-
ament tissue begins. There seems to be a temporal relationship
between the expression and deposition of the various matrix
molecules in ligaments, with the collagens (collagen 1 and 3)
being the major class of molecules that are deposited. In the
Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
normal MCL the ratio of collagen 1/collagen 3 is approxi-
mately 3/1, but in early scar tissue this ration is approximately
1/1.10,11 This ratio gradually becomes more normalized with
time, but the initial ratio of expression is certainly not
reflective of the make up of a normal ligament. Similarly,
expression of other normal matrix molecules in the healing
rabbit MCL such as the small leucine-rich proteoglycans
(SLRPs) including decorin, biglycan, lumican, and fibromo-
dulin is altered as compared with normal.12 Because these
molecules can serve various functions in these tissues such as
binding to collagens, growth factors, and growth factor re-
ceptors, their impact on healing tissues may be different from
their impact on normal tissues. Obviously, a number of other
molecules are also altered quantitatively and qualitatively from
normal tissue during the healing process, and many of these
alterations are initiated during the early deposition phase (also
called granulation tissue when an injury occurs in skin).
By extension, if the matrix deposited early during the
healing or repair process is altered compared with normal, the
organization of the repair tissue is also likely to be altered. It
has been shown that the organization of the matrix deposited
early following injury is disorganized compared with normal
tissue.4–6,13 Normal tissues are organized with respect to col-
lagen alignment and collagen fibril assembly, whereas collagen
expressed early following injury is not aligned and heteroge-
neous with regard to orientation in the tissue. Because the
latter is critical for function in a mechanically active environ-
ment such as a ligament, it is not surprising that the me-
chanical properties of the healing MCL ligament are severely
compromised compared with normal tissue.4,6 This lack of
matrix organization is not a tissue-specific trait, and has also
been reported at the site of injuries in other tissues, including
the granulation tissue of skin wounds.1
In addition to matrix deposition during this early phase
of healing, the cellularity of the repair tissue is increased
compared with normal1,7,8 and is more highly vascularized
than normal.14 The properties of the scar cells in the healing
MCL are different from normal cells11 and may result from
intrinsic differences or from differences caused by the growth
factor/cytokine rich environment. Some evidence from other
models has also implicated scar cells as being intrinsically
different, based on cell surface phenotype markers.15
Thus, early after injury there is an increase in cellularity
and a concomitant increase in vascularity to presumably sup-
port the nutritional needs of these cells. The increased vas-
cularity is likely caused by the release of angiogenic mediators
early after injury, which promotes the influx of new micro-
vessels. In some tissues that are fairly devoid of an endogenous
microvasulature, such as the meniscus, injuries to the most
avascular areas of the tissue do not heal well, whereas those
with a microvasculature do heal.16 Thus, the availability of a
microvasculature and an environment conducive to an angio-
genic response is critical for effective healing. As discussed in
other chapters in this issue,17 the early scar tissue is fairly
devoid of neural elements and thus any regulatory influences
of these elements on either the fibroblast-like cells or the micro-
vasculature would be absent.
Thus, the tissue deposited early after injury appears to be
an attempt to bridge the damaged area without regard to what
q 2005 Lippincott Williams & Wilkins
Basics of Wound Healing
was present before injury. Not only does this provisional matrix
have a different structural and cellular composition as com-
pared with normal tissue, but in the case of ligament injury,
this tissue is not necessarily even localized to the injury gap
but also may extend to surround the entire remaining ligament
midsubstance.4–6 This somewhat amorphous material, result-
ing from the initiation of an overt inflammatory response and
subsequent events is compromised at both the organizational
and functional levels, independent of whether it is a ligament,
tendon, or skin. Obviously, this early repair tissue should be
considered the ‘‘starting point’’ for further remodeling to better
approximate the functional needs of the tissue in its specific
environment. Whether such remodeling is effective depends in
part on the wound environment and patient factors. Also, the
clinical definition of ‘‘effective remodeling’’ depends on the
ultimate functional needs of the tissue (ie, MCL, ACL, flexor
tendon, skin). In some tissues, and in some cases (ie, chronic
skin wounds in an immunocompromised patient) wound
strength is a less important clinical outcome than is wound
closure. In contrast, in tissues such as ligament and tendon,
functionality is a very important factor.
The Remodeling Phase
The remodeling phase is a slow process and is
accompanied by alterations not only in matrix remodeling,
but also gene expression,7–9,12 cellularity,4,7,11 vascularity,14,17
and innervation.17 Thus, the scar tissue in a ligament such as
the MCL, or most other connective tissues, undergo a pro-
tracted process where the initially deposited material seems to
be turning over and the organization of collagen fibrils become
more oriented along the long axis of the ligament.13 In skin
wounds the converse is observed, whereby collagen fibril
orientation is more linearly arranged in the provisional matrix
and remodels to a cross-linked ‘‘basket-weave’’ arrangement,
providing structural strength in multiple directions. Further-
more, in pig models of skin wound healing, the mechanical
properties of the scar tissue does exhibit a directional polarity,
and is generally compromised compared with normal skin,
likely indicating incomplete remodeling (pers. comm.). Be-
cause the remodeling phase occurs slowly, and may take months
(ie, skin) or years (ie, tendon and ligament), the composition of
the scar or repaired tissue also changes with time, as does the
gene expression phenotype. Such changes are accompanied by
a decline in cellularity to near normal levels, as well as changes
in the vascularity and innervation.17,18 In the latter case, some
of the neuropeptides present early after injury are usually con-
sidered pro-inflammatory, whereas those appearing later are
considered more anti-inflammatory, so the potential regulatory
influence of neuropeptides on the outcome after injury is com-
plex and seems to be temporally regulated.
At the gene expression level, depending on the tissue
being examined, there is a gradual or abrupt decline in the
previously elevated levels that are apparent early after injury.
Some of this change is apparently gene- and tissue- dependent,
with differences between skin and ligaments being detected in
some models.7,8,12,19 In some pathologic conditions such as
joint contractures, even though the loss of range of motion has
stabilized, elevations in gene expression are maintained at
significantly high levels for a number of genes for a long
137
Hildebrand et al Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
period of time post-injury.20 Thus, a number of variables seem
to influence the rate of remodeling and the final outcome, and
it is not always possible to assign potential cause and effect
relationships.
Whereas many aspects of the repair tissue changes with
time post-injury, some aspects do not change in parallel. For
instance, collagen fibril diameters in normal adult ligaments
exhibit a biphasic distribution of diameters with primarily
small and large diameter fibrils, and a few medium diameter
fibrils.4,21 However, in the healing rabbit MCL, only small
fibrils are evident for over a year post-injury and even at
2 years post-injury the fibrils are still overwhelmingly small
diameter.21 In contrast, larger diameter fibrils appear in skin
fairly soon after injury, so there may be some tissue-specific
aspects to the remodeling phase of the healing process.
At the mechanical level, there are correlations between
aspects of matrix remodeling and the mechanical properties of
the repair or scar tissue with time post-injury. Thus, there is
a decline in the number of flaws or defects in the scar tissue
with time, as well as an increase in the collagen crosslink
density. Both are correlated with the increase in failure
strength of the repair tissue.4,22 However, even after protracted
time post-injury, the mechanical properties of a scar tissue in
a ligament such as the rabbit MCL is still compromised
compared with normal.4,19 Nevertheless, the scar tissue may be
functional for most activities even though it is not ‘‘normal’’.
This conclusion is also supported by the human experience
where athletes can again participate in many sports at a high
level after sustaining an injury to the MCL, the Achilles
tendon, or other similar structures. However, it remains to be
elucidated as to whether in such cases the injured tissue is
again functioning at a high level, or there has been some
compensatory increase in the functioning of other tissues (ie,
ligaments, musculotendinous units) in the motion segment or
organ such as the knee.23
LIGAMENT HEALING
Heterogeneity Between Ligaments
Whereas the above discussion focused on general
aspects of healing with some emphasis on ligament healing,
it should be pointed out that not all ligaments heal to the same
degree, and healing of ligament injuries seems to be influenced
by various factors including location (ie, extraarticular vs.
intraarticular), intrinsic aspects (which are largely unknown),
mechanical environment, as well as factors discussed in more
detail in the following sections.
An example of two ligaments that appear to heal dif-
ferently are the MCL and the ACL. The former is an
extraarticular ligament of the knee, and whereas the ACL is
also a knee ligament, it exists in an intraarticular environment.
In addition to location and environment, the two ligaments
also differ with respect to structure, mechanical or loading
environment, endogenous microvasculature, and some prop-
erties of the endogenous cells.4,24
As addressed earlier, the injured MCL of the healthy
young adult rabbit, or human, appears to heal quite well even
when the ligament is completely transected. Of course, this
138
ligament lies in an extraarticular environment where well-
vascularized tissues surround it and may keep the torn ends in
close proximity. Thus, the healing response initiated in this
environment can generate a large scar tissue that encompasses
both ends of the injured MCL. The large scar tissue mass
gradually remodels, likely under the influence of the me-
chanical environment, with the excess material not involved in
loading being resorbed. This leaves the resulting ligament-like
tissue with dimensions not unlike the original tissue. In this
situation, the injured MCL has surrounding tissues that can
serve as a surface on which to deposit the initial provisional
matrix that then remodels to ‘‘reform’’ a ligament-like structure.
In contrast, complete rupture of the intraarticular ACL
leads to the scenario where there is little surrounding tissue to
provide a vascular supply and keep the ends of the ACL in
close proximity. In the case of ACL injuries where the injury
occurs near the femoral insertion, the remainder of the ACL
can adhere to the PCL and form a ‘‘scar-like’’ material, but this
arrangement is functionally ineffective.
Recent studies in animal models has indicated that in
addition to the problem of having no template available after
complete rupture of the ACL, after a partial injury to the ACL
this ligament exhibits a healing response that is inferior to that
of the MCL.25 Thus in a sheep model, a complete MCL injury
was compared over time to a partial ACL injury where one
band of the ACL was transected. At the molecular level, the
initial healing response of the ACL was comparable to that of
the MCL at 6 weeks post-injury. However, by 12 weeks post-
injury, the MCL continued to heal whereas the healing process
in the ACL appeared not to be sustained and rather than to
improve in its properties, the ACL scar-like material appeared
to be undergoing a resorptive process and became less con-
solidated. Again, whether this was caused by the intraarticular
environment and a lack of mediators and cells available to
foster scar development and maturation, or the fact that the
scar in the ACL was somewhat ‘‘stress shielded’’ by the intact
band remaining, or a combination of factors, needs further
investigation. The cells available to the ACL to initiate repair
may be endogenous ACL cells, cells from the synovium, or
even mesenchymal stem cells in the synovial fluid. In contrast,
the cells available to the MCL may be very different; the
increased vascularity may supply more reparative cells in-
cluding mesenchymal stem cells from sites other than the
injury. Similarly, the ACL attempts to heal in the presence of
synovial fluid, which is rich in hyaluronans, factors that are
known to impact the functioning of many cells,26 whereas the
MCL is healing in a different milieu. Therefore, with so many
differences between the MCL and ACL healing environments,
it is difficult to assign cause and effect.
The possibility that the model used to assess the healing
of the ACL discussed above influenced the results caused by
stress shielding of the scar tissue emphasizes the point that
maturation of scar tissue in tissues that are designed to
function in a mechanically active environment likely means
that similar to normal ligaments and tendons, maturation of the
scar tissue requires mechanical loading to continue the re-
modeling phase of healing. Whereas some of this line of
thought will also be addressed in the chapter by Creighton and
Dahners regarding specific aspects of MCL healing, it is an
q 2005 Lippincott Williams & Wilkins
Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
important concept that needs reiteration as to its general ap-
plication to healing irrespective of tissue. Normal connective
tissues that function in a mechanically active environment
(actually most tissues) subscribe to the ‘‘use it or lose it’’
paradigm of tissue integrity.27 Increased loading leads to
adaptation, whereas decreased loading below a threshold leads
to atrophy. The same principle likely also holds for scar tissue
and immobilization beyond the initial phases of healing could
have a detrimental impact on outcome. Thus, stress-shielding
of scar tissue may have a detrimental effect on outcome
whether the injury is to a patellar tendon, a ligament, or even
skin. In fact, it could be argued that ‘‘immature’’ scar tissue
may be more susceptible to deprivation of loading than the
corresponding normal tissue because of the activation state of
the cells in the tissue. As mentioned elsewhere in this review,
some of the remodeling of an MCL scar and resorption of
excess scar tissue not apparently taking up load has been
noticed to occur between 6 and 14 weeks post-injury in an
unrestrained rabbit model.
Too much loading of a scar at too early a time point may
have a detrimental impact on the maturation of the scar. Thus,
recent studies using a rabbit model have used a gap injury to
the MCL, and after 6 weeks the ACL was transected. The
increased instability in the knee resulting from the loss of ACL
function lead to increased expression of inflammatory me-
diators in the healing MCL, likely caused by microinjuries to
the scar tissue, and resulted in a protracted healing response.28
Thus, there is a delicate balance between biology and me-
chanical environment when it comes to optimizing the basic
healing response in tissues such as ligaments, tendons, or skin.
Whereas not directly related to ligament healing, but
likely relevant to the discussion of ligament healing, is the
body of information regarding healing in environments lead-
ing to loss of function (ie, adhesion formation following abdom-
inal surgery or flexor tendon injuries). Thus, some tendon and
ligament injuries lead to formation of scar tissue that is
partially functional, but to regain as much function as possible
requires physiotherapy to ‘‘facilitate’’ the return to function
after the scar tissue has formed. The vigorous inflammatory
response associated with overt injury +/2 surgical interven-
tion to foster repair can lead to formation of adhesions, where
the ligament/tendon scar tissue is ‘‘bonded’’ to the surrounding
tissue and thus, such restrictions compromise function in
situations where movement is required. In situations such as
following flexor tendon injuries within the tendon sheath,
formation of adhesions can severely compromise function and
this is not readily corrected. Thus, if one could overcome the
apparent ‘‘deficiencies’’ of healing in the ACL where there is
minimal opportunity for adhesion formation compared with
the MCL, it may be possible to enhance healing without in-
creasing the risk of such side effects.
This line of thought also emphasizes the need to
minimize the induction of a vigorous inflammatory response in
some environments to assist in the repair process without side
effects such as adhesions. Because arthroscopy usually in-
duces less inflammation and tissue damage than open ap-
proaches, outcomes can be influenced and used to foster the
most optimal return to function. This line of discussion also
emphasizes the central role of the inflammatory response to
q 2005 Lippincott Williams & Wilkins
Basics of Wound Healing
injury discussed earlier in setting the stage for subsequent
events of the healing process (deposition of matrix, remodeling).
Healing is a Dynamic Process Leading to Repair
but Not Regeneration
Healing of ligaments and most other connective tissues
is a dynamic process that really has no ‘‘endpoint.’’ Successful
healing requires a complex and temporally regulated interplay
between different subsets of cells, mediators (cytokines,
growth factors, etc), protein cascades, and the physiology of
the host. The effectiveness of the process and the quality of the
outcome in an adult can be influenced by various factors, many
of which are discussed later. However, despite the impact of
the various factors, it is clear that the outcome is repair and not
regeneration in all soft connective tissues, except for muscle
and of course the hard tissue, bone. This point is emphasized
by writings from ships’ captains, from the 17th to 19th cen-
turies, whose men suffered from scurvy. Under conditions of
vitamin C deficiency, scars on men that had formed greater
than 20 years prior seemed to dissolve before normal skin was
affected, leaving gaping wounds where once there were scars.
Thus, even after many years, scar tissue is more ascorbate-
dependent than normal skin for maintenance of integrity in
humans. Whereas anecdotal, this point emphasizes the fact
that if we are to improve the outcome of scar formation and
maturation leading to regeneration, more information on the
regulation of normal healing is needed and the factors that
impact normal healing. Furthermore, information on how re-
generation is regulated in model systems (ie, animal models
where regeneration does occur) may yield further insights.
Such lines of investigation will provide critical information on
restrictive elements that may impact the outcomes in for in-
stance, in neonates versus adults, males versus females, liga-
ments versus skin.
FACTORS THAT INFLUENCE
HEALING OUTCOMES
Age
The healing process is influenced by age. It is obvious
from everyday life and research that incisional skin wounds
incurred at an early age appear to heal rapidly and without
residual scar. Because humans age, the same process following
a similar injury leads to obvious scars. Investigational studies
have revealed some of the basis for such changes and these
include alterations in inflammatory responses, a decline in
mesenchymal stem cell levels, a decline in the proliferative
capacity of fibroblast-like cells, metabolic changes, and changes
in the release or responsiveness to mediators such as growth
factors.29,30 Many parameters relevant to the complex healing
process change with aging, and simple solutions have been
promulgated from various sources, but it is unlikely that one
variable can correct what is apparently a complex set of factors.
Whereas much of the literature regarding the impact of aging
on healing has focused on skin, and it is well known that
even normal skin changes with age, it is also very likely that
parameters detected at the skin level also are relevant to
healing of other tissues such as ligaments. This conclusion is
139
Hildebrand et al
based on the fact that most of the earlier indicated variables
are system-specific rather than tissue-specific. In addition, it is
known that the biomechanical properties of ligaments and
tendons change with age (ie, become stiffer) because of accu-
mulated stresses and the incidence of injuries, and degen-
erative processes in many of these tissues increase with age
(aside from those associated with athletics).
One interesting area of research on age and the healing
response has led to some provocative findings. These are in the
area of fetal healing versus healing in the newborn. This is still
age-related, but obviously the environment that a fetus is in
versus that in which an independent newborn functions is
quite different, but the studies have pointed to some areas that
may prove to be fruitful in the future. The basis for this area
was the observation that wounds in some locations on a fetus,
because of surgical interventions, led to the apparent regen-
eration of the injured tissue rather than scar formation and
maturation.31,32 Such findings were primarily from incisional
skin wounds, and the ‘‘regeneration’’ phenotype was more readily
observed in some tissues than others. Further, the regeneration
phenotype was lost in the later stages of gestation. However,
investigation of differences in the healing process leading to
the regeneration phenotype versus scar formation has impli-
cated transforming growth factor-beta (TGF-beta) isoforms,
hyaluronic acid and the severity and extent of the inflam-
matory phase of healing in the differences. It is known that
inflammation is regulated differently in the fetus versus the
newborn, and in some species, the maturation of this regulation
differs. The finding of a possible role for TGF-beta3 in fetal
healing rather than TGF-beta1, which is the predominant form
involved in healing later in life, has led a number of groups to
investigate this avenue of research in detail.33–35 Manipulating
TGFbeta1 levels can modulate the scar phenotype. It remains
to be determined whether or not TGFbeta1 is the key or central
element in the phenotypic differences between fetal and adult
healing. It is most likely that the basis for the fetal phenotype is
multifactorial, and at least some of the steps identified may be
amenable to manipulation to improve healing of injured tissues
in the future.
Gender and Hormones
It is known in a variety of species that there is a sexual
dimorphism in many responses. Females tend to have a more
vigorous inflammatory response than males (discussed in
Ashcroft and Ashworth30), and as the inflammatory response
to injury is a central initial step in the healing process, this is a
consistent pattern of observation. It is also known that healing
responses in post-menopausal women decline separate from
the decline associated with aging. Application of estrogen can
reverse some of these deficiencies30 and it is believed to work
at the level of macrophages. Macrophages are inflammatory
cells that can mediate multiple functions in a wound site,
including serving as an important source of growth factors and
mediators key for effective wound healing.1,36 However, the
literature does not indicate that there is an overt alteration in
healing associated with the onset of puberty. Therefore, there
is either no direct impact of some hormones on the healing
response, or other variables operative in young individuals
override any hormonal influences and they only become more
140
Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
evident later in life such as post-menopause when other factors
have declined in an age-related manner.
Despite a paucity of information regarding hormonal
impact on wound healing in humans, some information is
available that supports the concept. First, in a rabbit model of
MCL healing, it has been shown that pregnancy affects the
metabolism of cells in the healing ligament (discussed in Hart
et al37). Pregnancy also affects the functioning of the normal
ligament (ie, laxity). However, pregnancy is associated with
changes in several hormones qualitatively and quantitatively,
and therefore it is not possible to ascribe the impact to any one
hormone. As normal ligament and joint function (ie, laxity)
can be influenced by the menstrual cycle in some women,38
such findings may indicate there is a role for hormones such as
estrogen, a hormone that can impact both macrophages and
ligament cells.39 However, whether normal alterations in
hormone levels associated with the menstrual cycle impact
wound healing in ligaments or other tissues remains a focus of
research in the author’s laboratories. One of the limitations of
current studies in animal models such as the rabbit is that these
animals are induced ovulators that do not cycle. However,
sexually mature pigs do cycle with hormonal variations some-
what similar to humans. Therefore, it should be possible to
address some of these issues directly in such models.
Genetics
It is clear that genetic factors play a role in some path-
ologic scarring or wound healing such as keloid formation.40
However, there is a paucity of information regarding a defini-
tive role for genetic factors in subtle or not so subtle variations
in ‘‘normal’’ healing. Whereas keloid formation is more com-
mon in some races than others (ie, dark skinned races), there is
no identifiable link between race and wound healing in other
tissues. There are polymorphisms associated with molecules
such as TGF-beta receptors,41 but they have not yet been linked
to wound healing phenotypes. There are some genetic links with
the functioning of connective tissues (ie, Marfan Syndrome,
benign Joint Hypermobility Syndrome),42,43 but we could not
find any links between such syndromes and wound healing
phenotype. If one assumes there is an important linkage in this
regard, it likely would have been discovered and reported by now.
However, it is possible that genetic factors or contribu-
tions could be obscured by other variables associated with
wound healing because of injury or surgical intervention. With
the availability of the human genome sequence, as well as
genome sequences for some species commonly used in wound
healing studies, genetic contributions to wound healing out-
comes may become more evident. It is apparent from talking
with orthopedic surgeons that there is a body of anecdotal
information that has implicated genetics in wound healing
following ligament injuries and surgical interventions, but it
does not yet have a defined scientific basis.
Whereas studies on wound healing in humans have not
yet elaborated an overt influence of genetics on this process,
recent studies in a porcine model of skin wound healing has
revealed a genetic component to healing outcome (pers.
comm.).7–9 That is, in the Yorkshire breed of pig, skin wound
healing proceeds through a series of defined molecular steps
similar to what has been described earlier,7,8 yielding a scar not
q 2005 Lippincott Williams & Wilkins
Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
unlike what occurs in humans. In contrast, wound healing in
the red Duroc pig yields a hypercontracted, hyperpigmented
scar, which has some similarities to human hypertrophic scar.9
The pattern of gene expression in this latter model is dis-
tinctively different from that detected in the Yorkshire model,
with a second ‘‘wave’’ of expression of molecules considered
proinflammatory being detected later in the healing process
(pers. comm.). More recent studies with Yorkshire x red Duroc
F1 pigs has revealed that the red Duroc phenotype is domi-
nant in such animals with a phenotype similar to the parental
breed with some minor differences noted. The offspring of F1
females crossed with a single Yorkshire male yielded back-
cross animals (BC) that exhibited a wound healing phenotype
similar to that observed in the Yorkshire parental breed (pers.
comm.). Thus, in pigs some evidence for involvement of
a genetic component in skin wound healing outcome has been
detected. Of course one major question remaining is whether
these findings are skin specific, or whether the skin results can
be extrapolated to wound healing in other tissues follow-
ing overt injury. Current studies are designed to explore the
answers to this and related questions using knee injury models
in pigs and healing in other organ systems (pers. comm.).
TISSUE ‘‘HISTORY’’
Whereas the impact of tissue history on the healing
outcome has not been the topic of intense investigation, it is
likely that the quality of the tissue prior to overt injury may
play a role in the wound healing process and the final outcome,
and therefore should be considered. Because of the paucity of
information, some of this discussion is admittedly speculative
and the relative importance of this variable on outcome is
largely unknown.
Mechanical History
Previously, it has been discussed that mechanobiology is
likely important in the healing outcome in tissues such as
ligaments, tendons, and related tissues. That is, depriving
healing ligaments of mechanical loading likely has a detri-
mental impact on healing outcome. Similarly, the question of
tissue quality and mechanical integrity of the tissue prior to
injury has not been addressed in a formal manner as far as
the present authors could detect. Intuitively, one may suspect
that because tissues such as skin, ligaments, and tendons can
adapt to increases in mechanical loading within a physiologic
window and decrease function when loading is consistently
decreased, the quality of the tissue prior to injury may also
impact the outcome. However, these issues need further inves-
tigation before conclusions can be made.
Biologic History
As discussed earlier, aging can likely influence the
healing outcomes. Also, the presence of previous injury, either
overt or subclinical, could also impact the healing outcome.
The healing outcome following re-injury could impact both
the quality of the outcome and the functioning of the healed
tissue. Majima et al28 have reported that re-injury of the acutely
healing MCL leads to alterations in gene expression of the
q 2005 Lippincott Williams & Wilkins
Basics of Wound Healing
healing tissue with re-expression of proinflammatory mole-
cules. However, the long-term consequence of such re-injury
on the mechanical properties of the tissue has not been
confirmed.
Because not all injuries to a tissue are overt, it is possible
that the accumulated cycles of injury and repair to a tissue
could impact the starting material following an overt injury. If
one extends this to the situation of a second acute injury, the
starting material following a second injury is really scar tissue
rather than normal tissue. This could impact the functional
outcome in at least 2 ways; first, the quality of the scar may be
compromised compared with the original scar tissue; and
second, the size of the scar may be increased and thus could
impact the functional outcome. Relevant to this point are the
results reported by Loitz-Ramage et al,44 which indicated that
the size of the wound and the resulting scar tissue has a
dramatic impact on the biomechanical outcome. Thus, surgical
interventions designed to limit or restrict the size of an injury
through suturing could improve the functional outcome after
ligament, tendon, or skin injuries.
Presence of Co-Morbidities
The presence of co-morbidities such as diabetes can
impact the healing outcome. Many patients with diabetes have
a compromised wound healing response due in part to an
impaired inflammatory response and elaboration of growth
factors. In addition, in conditions such as diabetes, the disease
could affect the quality of the connective tissue directly via
derivatization of the tissues and formation of advanced
glycation endproducts by carbohydrates.45
Co-morbidities could also impact the healing process
indirectly through the treatment interventions designed to
control the co-morbidities. Examples of this scenario include
rheumatoid arthritis patients treated with biologics such as
those designed to neutralize inflammatory components (Remicade
and Embril directed at TNF, IRAP directed at Il-1). In various
inflammatory diseases, patients are treated with glucocorti-
coids. Glucocorticoids have profound effects on inflammatory
processes,46 and recent studies have shown that exposure to
systemic exogenous glucocorticoids during ligament healing in
a rabbit model impacts gene expression quite differently than
in normal tissue (pers. comm.). Whether such exposure has
a long term impact on functional outcomes is not known. In
a rabbit model, a single local glucocorticoid treatment of a
healing ligament resulted in reduced biomechanical properties
of the scar, possibly because of a delay in the maturation/re-
modeling of the healing tissue.47,48 However, in a rat model,
exposure to glucocorticoids via a direct injection did not overtly
influence the mechanical outcome.49 Furthermore, Beer et al50
have reported that exposure of excisional skin wounds in mice
to glucocorticoids leads to alterations in gene expression in the
scar tissue. Whether the differences between the impact of
glucocorticoids on skin wounds50 and ligament wound sites
(pers. comm.) are because of tissue or species differences
remains to be clarified.
Approaches to Improve Healing Outcomes
From earlier in this discussion, it is readily apparent that
wound healing in the adult under the most optimal conditions
141
Hildebrand et al
should be considered tissue repair not regeneration. The ef-
fectiveness of this repair is dynamic as a person ages and the
outcome can be influenced by several variables or factors. For
tissues like skin, the outcome in the healthy individual is
usually functionally satisfactory, but it may be less than ideal
from a cosmetic perspective. For other tissues like a ligament
or tendon, the mechanical outcome may be less than ideal,
depending on the expectations of tissue use post-injury and the
occurrence of side-effects such as adhesions. Thus, there is
a need to better understand the critical basic steps in normal
wound healing and develop strategies and procedures to
improve healing outcomes. Ideally, the former would precede
the latter, but the reality is that some of the research in this area
has advanced in parallel, in part, because some approaches are
restricted to some species and the multistep nature of the
healing process likely has several critical steps at each of the
phases discussed earlier.
Improved Understanding of Critical Steps
Thorough Genetic Manipulation
It is apparent that there are several ways to improve our
understanding of critical steps in the healing process. One is to
understand the basis for ‘‘accidents’’ of nature in patients, such
as those with keloid formation or those with other genetic
influences on outcome. However, the incidence of the latter is
sometimes very rare. A second approach is to understand
impaired healing associated with co-morbidities and the basis
for deficiencies, such as in diabetes. A third approach has been
offered by the availability of techniques to genetically alter
animals such as mice through either knocking out genes, or
inserting genes into the germ line and thus impact healing. The
advantage of approaches one and two is that they are in
humans, but may occur rarely. The advantage of the third
approach is that it is powerful and several genes can be
influenced in a programmed manner (either completely, or in
a tissue-specific manner). However, the disadvantage of this
approach is that mice are loose skinned animals and their skin
heals by contraction rather than re-epithelization as in humans.
In addition, it is difficult to functionally assess some of the
tissues in mice because of their small size. Thus, the question
of whether the findings in mice can be directly extrapolated to
humans is always present and there are some limitations.
A complete list of the knockout and transgenic mice that
have contributed to our understanding of critical steps in the
process of wound healing in mice is beyond the scope of this
review. However, several reviews on the subject, including
those by Wener and Gross51 and Arbeit and Hirose,52 are
available. The bottom line of this approach is that multiple
important steps have been identified, as might be expected
given the complexity of the complete process and the temporal
relationships that have to happen in sequence. However, even
though the results of such studies may not be directly extrapo-
lated to other species such as humans, they do offer some
direction for further studies in other species and therefore the
process of understanding can be hopefully accelerated.
Application of Exogenous Growth Factors
From earlier in the discussion, a well-regulated growth
factor ‘‘cascade’’ is important for the orderly progression of
142
Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
steps in the wound healing process. This has been char-
acterized in multiple experimental systems and in humans with
healing defects (ie, patients with diabetes). In pig models,
among others, it has been shown that expression of some
growth factors (ie, TGF-beta) is required for the subsequent
expression of other factors, whereas some such as CTGF can
autoregulate their own expression.53,54 Therefore, because of
the central role of growth factors in the process, several studies
have explored the use of exogenous growth factors to ‘‘improve’’
the healing outcome in tissues like skin, ligaments and
tendons.55 Whereas some studies have resulted in improve-
ments in healing outcomes, others have not. For instance,
exposure of healing ligaments in a rabbit model to exogenous
TGFbeta1 led to larger scars, but no improvement in the
mechanical outcome.56 Because this study was performed in
healthy adult rabbits, it is likely that endogenous levels of
TGFbeta were sufficient and adding more would not lead to
improvements in outcomes.
In contrast, many individuals with diabetes exhibit
impaired wound healing and can develop chronic wounds that
do not readily heal. Because patients with diabetes can exhibit
impaired inflammatory responses, some of the growth factors
that have been promoted to improve healing are associated
with the early phases of healing. Other growth factors may also
be of benefit in situations such as diabetes, possibly as a
combination therapy.
Application of Stem Cells
Recent studies have indicated that mesenchymal stem
cells (MSC) exist and can participate in the healing of injured
connective tissues (discussed in57). Such cells are circulating
in the body, but their levels decline with age. As such cells can
differentiate under the influence of growth factors and other
mediators, the potential exists that they could be enticed to
assume the phenotype of the endogenous cells in an injury site
and contribute to the ‘‘regeneration’’ of the injured tissue or at
least push the scar tissue further down the path to regeneration.
The key in this regard is to identify the appropriate steps to
achieve this goal. Alternatively, such cells may assist in the
healing of injuries in individuals that have compromised
healing abilities caused by the presence of co-morbidities or
aging. Because such cells can be isolated from young people
and stored frozen until needed, autologous cells could be used
and thus avoiding the complications and risks of allogenic
cells. At this time, the key issues surrounding MSC are
focused on their propagation, delivery to needed sites, and the
conditions required to induce their differentiation optimally
and effectively. Once many of these issues are elucidated, their
potential in enhancing wound healing outcomes may be
realized at some point in the future.
Gene Therapy Approaches
The use of gene therapy approaches to improve healing
outcomes or correct other deficiencies has had a ‘‘checkered’’
past. In part, this has resulted from the targets chosen, the
vectors used to deliver the gene, the ability to deliver the gene
to the cells, and the efficiency of the expressed target to
influence outcomes. Some studies have tried to deliver the
gene construct directly to the injury site, whereas others have
q 2005 Lippincott Williams & Wilkins
Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
transfected cells in vitro and then delivered them to the wound
site.57–59 In this regard, transfecting MSC in the future may be
a real possibility, once the best genes are identified and the
most optimal circumstances for their use have been identified.
However, it is clear that some gene therapy approaches
have lead to increased understanding of the role of specific
steps in healing, and that approaches to manipulate gene
expression both in vitro and in vivo can provide valuable
information. Some of the newer approaches such as the use of
siRNA to silence genes54 are very useful for research purposes,
but they are a long way from the clinical realm until we better
understand the consequences of their application.
Generation of Replacement Tissues
One approach that could be anticipated to improve
healing of some tissues is to generate a replacement tissue that
is similar or identical to the endogenous tissue prior to injury.
For instance, in the case of the ruptured ACL, one could
replace the damage tissue with a replacement that has the
properties and function of the original. We have certainly not
yet achieved this goal, and have not been able to either gen-
erate a tissue replacement that mimics the original, nor is it
incorporated into the injury site as the original. Whereas this
topic is the subject of another review in this issue, tissue
engineering offers the potential to allow for restoration of
a tissue. At this time, several issues remain from the per-
spective of the tissue generated and the incorporation of the
engineered tissue once it is implanted. Obviously, tissues such
as skin pose a different set of problems than does a ligament or
tendon, but there are some common elements and hopefully
this approach will come to fruition in the near future for some
applications.
CONCLUSION AND FUTURE DIRECTIONS
It is clear from the earlier discussion that considerable
progress in our understanding of wound healing and the
factors that can influence outcomes has been accomplished
over the past few decades. In part, some of this progress has
been the result of integration of research perspectives from
many traditionally different expertise domains and an appre-
ciation of the role of some newer findings and the application
of newer technologies to the problems associated with wound
healing. An example of the former is the role of systemic and
local MSC in repair processes, and an example of the latter is
molecular techniques such as real time quantitative RT-PCR
and related methodology.
Whereas considerable progress has been made, it is
obvious that there are gaps in our knowledge base, and many
challenges remain. These include a better knowledge of the
development and maturation of function of normal tissues (so
as to better understand what the goal actually is), improved
understanding of the regulation of inflammatory responses,
which set the stage for subsequent healing processes, delin-
eating how local environmental factors modulate the healing
process, better characterization of the regulation of the
interface between biology and biomechanics (ie, mechanobi-
ology) and how it affects the functionality of the reparative
outcome, and finally, determining how genetics and genomics
q 2005 Lippincott Williams & Wilkins
Basics of Wound Healing
impact the functioning of the generalized healing process.
Successfully addressing these and related issues will all likely
require multidisciplinary approaches, the availability of good
experimental models, and the effective translation of in-
formation from other domains to the area of wound healing.
ACKNOWLEDGMENTS
Studies from the author’s laboratories were supported
by grants from the Canadian Institutes of Health Research,
particularly the Institute for Gender and Health, The Arthritis
Society, the Canadian Arthritis Network, and the Canadian
Space Agency. DAH is the Calgary Foundation-Grace Glaum
Professor in Arthritis Research. CLG-B was supported by an
Industry-NSERC Studentship and ASK by a MD/PhD Student-
ship from CIHR. KAH is an AHFMR Clinical Investigator and
also supported by the Health Research Foundation. The
authors acknowledge that representative citations in many
areas were used, and apologize to those individuals whose
studies were not cited.
REFERENCES
1. Witte MB, Brabul A. General principles of wound healing. Surg Clin
North Am. 1997;77:509–528.
2. Diegelmann RF, Evans MC. Wound healing: an overview of acute, fibrotic
and delayed healing. Front Biosci. 2004;9:283–289.
3. Goldman R. Growth factors and chronic wound healing: past, present, and
future. Adv Skin Wound Care. 2004;17:24–35.
4. Hildebrand KA, Hart DA, Rattner JB, et al. Ligament injuries:
pathophysiology, healing and treatment considerations. Musculoskeletal
Rehabilitation. Philadelphia, Elsevier Science.(in press)
5. Frank CB, Amiel D, Akeson WH. Healing of the medial collateral
ligament of the knee. A morphological and biochemical assessment. Acta
Orthop Scand. 1983;54:917–923.
6. Shrive NG, Thornton GM, Hart DA, et al. Knee ligaments: structure,
function, injury and repair. Philadelphia: Williams and Wilkins. 2002;97–
112.
7. Wang JF, Olson ME, Reno CR, et al. Molecular and cell biology of skin
wound healing in a pig model. Connect Tissue Res. 2000;41:195–211.
8. Wang JF, Olson ME, Reno CR, et al. The pig as a model for excisional
wound healing: characterization of the molecular and cellular biology, and
bacteriology of the healing process. Comp Med. 2001;51:341–348.
9. Gallant CL, Olson ME, Hart DA. Molecular, histologic, and gross
phenotype of skin wound healing in red Duroc pigs reveals an abnormal
healing phenotype of hypercontracted, hyperpigmented scarring. Wound
Repair Regen. 2004;12:305–319.
10. Murphy PG, Loitz BJ, Frank CB, et al. Influence of exogenous growth
factors on synthesis and secretion of collagen types I and III by explants of
normal and healing rabbit ligaments. Biochem Cell Biol. 1994;72:403–
409.
11. Murphy PG, Frank CB, Hart DA. The Anterior Cruciate Ligament. New
York: Raven Press; 1993:pp.165–177.
12. Boykiw R, Sciore P, Reno C, et al. Altered levels of extracellular matrix
molecule mRNA in healing rabbit ligaments. Matrix Biol. 1998;17:371–
378.
13. Liu ZQ, Rangayyan RM, Frank CB. Statistical analysis of collagen
alignment in ligaments by scale-space analysis. IEEE Trans Biomed Eng.
1991;38:580–588.
14. Bray RC, Rangayyan RM, Frank CB. Normal and healing ligament
vascularity: a quantitative histological assessment in the adult rabbit
medial collateral ligament. J Anat. 1996;188:87–95.
15. Koumas L, Smith TJ, Feldon S, et al. Thy-1 expression in human
fibroblast subsets defines myofibroblastic or lipofibrobalstic phenotypes.
Am J Pathol. 2003;163:1291–1300.
16. Kobayashi KK, Fujimoto E, Deie M, et al. Regional differences in the
healing potential of the meniscus-an organ culture model to eliminate the
143
Hildebrand et al
influence of the microvasculature and the synovium. Knee. 2004;11:271–
278.
17. Bray RC, Salo PT, Lo IK, et al. Normal ligament structure, physiology and
function. Sports Med Arthrosc Rev.(this issue)
18. Ackermann PW, Li J, Lundeberg T, et al. Neural plasticity in relation to
nociception and healing of rat Achilles tendon. J Orthop Res. 2003;21:
432–441.
19. Frank CB, Hart DA, Shrive NG. Molecular biology and biomechanics of
normal and healing ligaments. Osteoarthritis Cartilage. 1999;7:130–140.
20. Hildebrand KA, Zhang M, Hart DA. Evidence for a high rate of matrix
turnover in chronic human elbow contractures. Clin Orthop. (accepted for
publication).
21. Frank CB, McDonald D, Shrive NG. Collagen fibril diameters in the rabbit
medial collateral ligament scar: a longer term assessment. Connect Tissue
Res. 1997;36:261–269.
22. Frank CB, McDonald D, Wilson J, et al. Rabbit medial collateral ligament
scar weakness is associated with decreased collagen pyridinoline crosslink
density. J Orthop Res. 1995;13:157–165.
23. Frank CB, Shrive NG, Boorman RS, et al. New perspectives on
bioengineering of joint tissues: joint adaptation creates a moving target for
engineering replacement tissues. Ann Biomed Eng. 2004;32:458–465.
24. Frank CB, Jackson DW. The science of reconstruction of the anterior
cruciate ligament. J Bone Joint Surg Am. 1997;78:1556–1576.
25. Lo IK, Marchuk LL, Timmerman SA, et al. A comparison of early healing
of ovine medial collateral ligament and anterior cruciate ligament injuries:
a multidisciplinary approach. Am J Sports Med. Submitted/Dec 2004.
26. Brown JA. The role of hyaluronic acid in wound healing’s proliferative
phase. J Wound Care. 2004;13:48–51.
27. Hart DA, Natsuume T, Sciore P, et al. Mechanobiology: similarities and
differences between in vivo and in vitro analysis at the functional and
molecular levels. Recent Res Devel Biophys Biochem. 2002;2:153–177.
28. Majima T, Lo IK, Randle JA, et al. ACL transaction influences mRNA
levels for collagen type I and type III and TNF-alpha in MCL scar.
J Orthop Res. 2002;20:520–525.
29. Gosain A, DiPetro LA. Aging and wound healing. World J Surg. 2004;28:
321–326.
30. Ashcroft GS, Ashworth JJ. Potential role of estrogens in wound healing.
Am J Clin Dermatol. 2003;4:737–743.
31. Mast BA, Diegelmann RF, Krummel TM, et al. Scarless wound healing in
the mammalian fetus. Surg Gynecol Obstet. 1992;174:441–451.
32. Colwell AS, Longaker MT, Lorenz HP. Fetal wound healing. Front Biosci.
2003;8:s1240–s1248.
33. Hu M, Sabelman EE, Cao Y, et al. Three-dimensional hyaluronic acid
grafts promote healing and reduce scar formation in skin incision wounds.
J Biomed Mater Res B Appl Biomater. 2003;67:586–592.
34. Liu W, Wang DR, Cao YL. TGF-beta: a fibrotic factor in wound scarring
and a potential target for anti-scarring gene therapy. Curr Gene Ther.
2004;4:123–136.
35. Ferguson MW, O’Kane S. Scar-free healing: from embryonic mechanisms
to adult therapeutic intervention. Philos Trans R Soc Lond B Biol Sci.
2004;359:839–850.
36. Ashcroft GS, Mills SJ, Lei K, et al. Estrogen modulates cutaneous wound
healing by down regulating macrophage migration inhibitory factor.
J Clin Invest. 2003;111:1309–1318.
37. Hart DA, Reno C, Frank CB, et al. Pregnancy affects cellular activity but
not tissue mechanical properties, in the healing rabbit MCL. J Orthop Res.
2000;18:462–471.
144
Sports Med Arthrosc Rev  Volume 13, Number 3, September 2005
38. Deie M, Sakamaki Y, Sumen Y, et al. Anterior knee laxity in young
women varies with their menstrual cycle. Int Orthop. 2002;26:154–156.
39. Seneviratne A, Attia E, Williams RJ, et al. The effect of estrogen on ovine
anterior cruciate ligament fibroblasts: cell proliferation and collagen
synthesis. Am J Sports Med. 2004;32:1613–1618.
40. Marneros AG, Norris JE, Olsen BR, et al. Clinical genetics of familial
keloids. Arch Dermatol. 2001;137:1429–1434.
41. Watanabe Y, Kinoshita A, Yamada T, et al. A catalog of 106 single-
nucleotide polymorphisms (SNPs) and 11 other types of variations in
genes for transforming growth factor-beta1 (TGF-beta1) and its signaling
pathway. J Hum Genet. 2002;47:478–483.
42. Zweers MC, Hakim AJ, Grahame R, et al. Joint hypermobility syndromes:
the pathophysiologic role of tenascin-X gene defects. Arthritis Rheum.
2004;50:2742–2749.
43. Loeys BL, Matthys DM, dePaepe AM. Genetic fibrillinopathies: new
insights in molecular diagnosis and clinical management. Acta Clin Belg.
2003;58:3–11.
44. Loitz-Ramage BJ, Frank CB, Shrive NG. Injury size affects long-term
strength of the rabbit medial collateral ligament. Clin Orthop. 1997;337:
272–280.
45. Ahmed N. Advanced glycation endproducts-role in pathology of diabetic
complications. Diabetes Res Clin Pract. 2005;67:3–21.
46. Amstead GM. Steroids, retinoids, and wound healing. Adv Wound Care.
1998;11:277–285.
47. Wiggins ME, Fadale PD, Barrach MG, et al. Healing characteristics of
a type I collagenous structure treated with corticosteroids. Am J Sports
Med. 1994;22:279–288.
48. Walsh WR, Wiggins ME, Fadale PD, et al. Effects of delayed steroid
injection on ligament healing using a rabbit medial collateral ligament
model. Biomaterials. 1995;16:905–910.
49. Campbell RB, Wiggins ME, Canistra LM, et al. Influence of steroid
injection on ligament healing in the rat. Clin Orthop. 1996;332:242–253.
50. Beer HD, Fassler R, Werner S. Glucocorticoid-regulated gene expression
during cutaneous wound repair. Vitam Horm. 2000;59:217–239.
51. Werner S, Grose R. Regulation of wound healing by growth factors and
cytokines. Physiol Rev. 2003;83:835–870.
52. Arbeit JM, Hirose R. Murine mentors: transgenic and knockout models of
surgical disease. Ann Surg. 1999;229:21–40.
53. Wang JF, Olson ME, Ball D, et al. Recombinant connective tissue growth
factor modulates porcine skin fibroblast gene expression. Wound Repair
Regen. 2003;11:220–229.
54. Wang JF, Olson ME, Ma LL, et al. Connective tissue growth factor siRNA
suppresses porcine skin fibroblast gene expression. Wound Repair Regen.
2004;12:205–216.
55. Woo SL, Smith DW, Hildebrand KA, et al. Engineering the healing of the
rabbit medial collateral ligament. Med Biol Eng Comput. 1998;36:359–
364.
56. Hildebrand KA, Hiraoka H, Hart DA, et al. Exogenous TGF-beta1 alone
does not improve early rabbit MCL healing in vivo. Canad J Surg. 2002;
45:330–336.
57. Hart DA, Shrive NG, Goulet F. Tissue engineering of ACL replacements.
Sports Med Arthrosc Rev.(this issue)
58. Lo IK, Randle JA, Majima T, et al. New directions in understanding and op-
timizing ligament and tendon healing. Curr Opin Orthop. 2000;11:421–428.
59. Hildebrand KA, Frank CB, Hart DA. Gene intervention in ligaments and
tendons: current status, challenges and future directions. Gene Ther. 2004;
11:368–378.
q 2005 Lippincott Williams & Wilkins