Clin Plastic Surg 30 (2003) 109 – 126
Pathophysiology of nerve injury
Sergio P. Maggi, MD, James B. Lowe III, MD, MBA,
Susan E. Mackinnon, MD*
Division of Plastic and Reconstructive Surgery, Washington University School of Medicine, Suite 17424 East Pavilion,
One Barnes-Jewish Hospital Plaza, St. Louis, MO 63110, USA
Anatomy
Connective tissue
A normal myelinated nerve is represented in Fig. 1,
which demonstrates the extensive connective tissue
that makes up the major component of the nerve and
provides the scaffolding for the axons and their
Schwann cells. Early studies of peripheral nerve anatomy delineated much of the fundamental architecture.
Key and Retzius [1] detailed the subdivision of epineurium, perineurium, and endoneurium. The loose
areolar tissue external to the epineurium is termed the
mesoneurium. It is continuous with the epineurium
and is critical to the longitudinal excursion of peripheral nerves. Millesi [2] stressed the importance of
the mesoneurium in aiding the movement of the nerve
in its surrounding tissue bed. Millesi [3] has shown that
the median nerve moves up to 9.6 mm with wrist
flexion and slightly less with extension. Wilgis and
Murphy [4] demonstrated that the median and ulnar
nerves glide longitudinally approximately 7.3 mm and
9.8 mm, respectively, with elbow flexion. Nerve
excursion is critical in normal physical activity, and
chronic nerve compression or associated fibrosis may
significantly impair the ability of a nerve to glide,
resulting in pain or discomfort.
The outer sheath of the nerve is the external
epineurium and is composed primarily of collagen
and elastic fibers. The internal epineurium is the
structure that invests the fascicles, which are composed of myelinated and unmyelinated nerve fibers
[5]. Fascicles vary in quantity and in size; much of
the variety depends on the anatomic level and position of the peripheral nerve. Fascicular anatomy of a
peripheral nerve can consist of a single fascicle
(monofascicular), few fascicles (oligofascicular), or
many fascicles (polyfascicular) (Fig. 2).
Each fascicle is encircled by the connective tissue
known as perineurium. The cellular make-up of the
perineurium consists of collagen fibrils dispersed
among perineural cells [6]. The perineural layer is a
major site for selective permeability because of the
tight junctions that form between adjacent perineural
cells. Surrounding each individual axon is a loose
gelatinous collagen matrix called the endoneurium
[7]. Although the endoneurial tissue has some tensile
strength, the perineurium provides the peripheral nerve
with most of its tensile strength and elasticity [8].
The volume and quality of intra- and extraneural
connective tissue varies among different nerves and
within the same nerve at different levels and anatomic
sites [5]. For instance, superficial nerves have been
shown to have substantial increases in connective
tissue as they cross joint surfaces. The sciatic nerve
has been shown to consist of approximately 87%
connective tissue as it crosses the hip, which may be
the result of a local reaction to repetitive strain or as a
protective mechanism of nature [5].
Neural topography
* Corresponding author.
E-mail address: [email protected]
(S.E. Mackinnon).
In the more proximal aspects of the extremities,
there is a significant plexus formation between the
fascicles, but the sensory and motor topography is
0094-1298/03/$ – see front matter D 2003, Elsevier Science (USA). All rights reserved.
doi:10.1016/S0094-1298(02)00101-3
110
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
Fig. 1. Diagrammatic representation of the cross-section of a normal peripheral nerve demonstrating the connective tissue and
nerve tissue components. (Adapted from Mackinnon SE, Dellon AL. Surgery of the peripheral nerve. New York: Thieme, 1988.
p. 36; with permission.)
specific. In the distal aspects of the extremity, the
plexus formation between fascicles is far less extensive. Sunderland [9] reported that the internal topography of peripheral nerves became increasingly more
organized as they progressed distally (Fig. 3).
Jabaley [10] has shown that nerve fascicles have
significantly fewer interfascicular connections, or
plexus formation, distally as compared with proximally in the extremity.
Jabaley [10] demonstrated that the recurrent motor
branch of the median nerve can be readily dissected
without plexus formation for approximately 70 mm,
and the anterior interosseous nerve at the level of the
medial humeral epicondyle can be separated proximally approximately 60 mm. The development of
more functionally distinct fascicles in distal peripheral nerves allows a more extensive surgical dissection via internal neurolysis and provides a variety of
reconstructive options (Fig. 4). Therefore, an understanding of the anatomical patterns and topography of
different peripheral nerves is an essential component
to peripheral nerve surgery [11].
Blood supply
Fig. 2. Schematic diagram of the three types of fascicular
patterns: monofascicular (left top), oligofascicular (center
top), and polyfascicular (right top). Below are schematic
examples of unmyelinated (left) and myelinated axons (right).
A single Schwann cell can envelop multiple unmyelinated
axons or concentrically wrap its cell membrane around an
axon to form a myelinated fiber. (From Winagrad J,
Mackinnon S. Peripheral nerve injuries: repair and reconstruction. In: Mathis S, Hentz VR, editors. Plastic surgery.
Philadelphia: Elsevier and Saunders; 2003; with permission.)
The blood supply to peripheral nerves occurs via
an intrinsic and extrinsic system. The intrinsic system
circulates within the epineural sheath, and the
extrinsic circulation is derived from small vessels
outside of the nerve. These vessels can enter the
nerve via one of three patterns: segmental with no
dominant pedicle, one dominant pedicle coursing
longitudinally with the nerve, and multiple dominant
pedicles along the length of the nerve [12] (Fig. 5).
The external vessels enter the mesoneurium and
communicate with the epineurial space via the vasa
nervorum. At this junction, a plexus formation develops and runs longitudinally within the perineurial
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
111
rabbit sciatic nerve model to evaluate intraneural
blood flow. He demonstrated that a peripheral nerve
could be raised independent of its extrinsic inflow
without evidence of ischemic damage. He demonstrated that a nerve receiving flow from only one extrinsic
blood vessel maintains perfusion over a diameter/
length ratio of 1:45 without evidence of ischemic
changes. This knowledge is useful when a peripheral
nerve needs to be extensively mobilized (eg, submuscular transposition of the ulnar nerve) and supports the
theory of a robust and hardy intrinsic vascular system.
Blood-nerve barrier
A barrier system exists in the peripheral nerve that
is an extension of the blood-brain barrier. The bloodnerve barrier is composed of the internal layers of the
perineurial sheath and the tight junction of the endothelial cells of the endoneurial microvessels. These
tight junctions regulate the intraneural environment
and provide a level of immunologic protection. Smith
et al [16] showed that the blood-nerve barrier is not
functional before 13 days of life because clefts are
present between the endothelial cells. After 16 days,
these clefts are replaced by tight junctions, resulting
in a blood-nerve barrier.
Nerve injury can result in a breakdown in the
blood-nerve barrier at the level of the microvessels,
resulting in an increase in pressure within the fascicle.
The lack of a lymphatic circulation within the endoneurial space and subsequent unresolved edema
within the compartment can interfere with the microcirculation of the fascicles [17]. Leaking capillaries
within the endoneurium permits the eventual accumulation of fluid and proteins. As the pressure rises
within the endoneurial space, a ‘‘mini compartment’’
syndrome can develop, resulting in ischemic damage
to that portion of the nerve.
Nerve fibers
Fig. 3. Sunderland’s classic reconstruction of the internal
topography of a peripheral nerve with marked plexus
formation. (From Sunderland S. Nerves and nerve injuries.
Edinburgh: Churchill Livingstone; 1978; with permission.)
space. This plexus enters the endoneurium at an
oblique angle to anastomose with the intrinsic circulation that surrounds each nerve fascicle. This
junction is a site of potential circulatory compromise
with increase in endoneurial pressure.
Many studies have shown that peripheral nerves
have a well-developed intrinsic blood supply [13 – 15].
Breidenbach [12] used radioactive microspheres in the
The basic neural structure is made up of axons and
their accompanying Schwann cells. Nerve fibers are
myelinated or unmyelinated. Sensory and motor
nerves contain both types of fibers in a ratio of
4 unmyelinated to 1 myelinated. Unmyelinated fibers
constitute 75% of cutaneous nerves and 50% of motor
nerves. They are composed of several axons encircled
by a single Schwann cell. In most cases, unmyelinated
axons are small, with diameters averaging 0.15 to
2.0 mm. The axons of myelinated nerve fibers are
individually enveloped by a single Schwann cell
(Fig. 6). Histologically, the myelinated fiber consists
of a center core of cytoplasm (axoplasm) surrounded
112
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
Fig. 4. (A) A cadaver dissection demonstrating the internal fascicles of the median nerve with the hand to the right of the page.
(B) A closer view of the intraneural topography at the proximal arm above the elbow showing marked plexus formation between
fascicles. (C) A closer view of the intraneural topography just distal to the elbow with less plexus formation. (D) A closer view of
the intraneural topography at the wrist where little to no plexus formation exists. (From Mackinnon SE, Dellon AL. Surgery of
the peripheral nerve. New York: Thieme, 1988. p. 7 – 8; with permission.)
by a confluent membrane (axolemma) that is further
surrounded by the Schwann cells laminar myelin
sheath. The Schwann cells act as specialized satellite
cells that can be identified definitively at the electromicroscopic level by their double basement membrane
[18,19].
At the junction between adjacent Schwann cells,
the axolemma becomes exposed at a gap known as
the node of Ranvier. This gap is where sodium
channels cluster and are found regularly along the
axon. The nodes of Ranvier aide in the propagation of
the depolarizing action potentials along the length of
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
113
Fig. 4 (continued ).
the nerve. The myelinated portions of the axon act as
insulated areas that accelerate axonal conduction as
the depolarization wave ‘‘jumps’’ from node to node.
This is referred to as saltatory conduction, and it
increases the speed of transmission more than would
be accomplished by simply increasing axonal diameter. Unmyelinated fibers are able to propagate
electrical impulses at speeds of 2 to 2.5 m/s. Myelinated fibers, by contrast, propagate impulses at
speeds ranging from 3 to 150 m/s [20,21].
Axons originate from their corresponding cell
bodies, which are located in the spinal cord, dorsal
root ganglion, and autonomic ganglia. Sensory nerve
fibers originate from the cell bodies in the dorsal root
ganglia and join other spinal nerves to connect with
sensory end organs. Motor nerve fibers (along with
presynaptic sympathetic ganglia) arise from the
anterior horn of the spinal cord and terminate at the
neuromuscular junction in the muscle. The presynaptic sympathetic ganglia send their axons out to the
postsynaptic sympathetic ganglia in the periphery,
which then go on to innervate blood vessels, skin,
and hair follicles.
The majority of a neuron’s cytoplasm is included
in the volume of the axon [22]. Ducker’s [22,23]
analogy is pertinent: ‘‘ If the central cell body were
the height of an average man, it’s axon would be one
or two inches in width and would extend more than
two miles.’’ The diameter of nerve fibers correlates
with the velocity of transmission and function. Nerve
fibers have been classified by variation in diameter
and speed. Group A fibers are the largest and fastest
conducting fibers. They are mostly myelinated
and take pulses in efferent and afferent directions.
Group B fibers contain myelinated autonomic and
preganglionic efferent fibers. Group C fibers are the
smallest and slowest. They are usually nonmyelinated
and serve in the transmission of thermoreceptive and
interoceptive signals.
Fig. 5. The blood supply of a peripheral nerve classified into
three patterns. (Left) Segmental supply with no dominant
pedicle. (Center) A single dominant pedicle running
longitudinally. (Right) Multiple dominant pedicles. (From
Winagrad J, Mackinnon S. Peripheral nerve injuries: repair
and reconstruction. In: Mathis S, Hentz VR, editors. Plastic
surgery. Philadelphia: Elsevier and Saunders; 2003;
with permission.)
Axoplasmic transport
Axoplasmic transport can be divided bidirectionally into an antegrade and retrograde system. The
transport process occurs via fast or slow transport in
the antegrade direction and via a fast system in the
retrograde direction. Axoplasmic transport maintains
114
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
Fig. 6. Transverse section (uranyl acetate and lead citrate, 13,000) demonstrating a Schwann cell nucleus with an associated
double basement membrane.
the structure of the nerve and provides the functional
capacity of neurotransmitters. Axonal transport is
energy dependent and can be disturbed by systemic
derangements and direct nerve injury. In 1941, it was
noted that the poliovirus infections used retrograde
axonal transport to reach the anterior motor horn [24].
Subsequent work by Droz and Leblond [25], using
labeled amino acids, determined that the antegrade
transport of neurotransmitter vesicles along the nerve
occurred at a rate of 1 and 1.5 mm per day.
Slow antegrade flow (1 to 6 mm per day) is used
for the transport of major neural building blocks such
as actin, tubulin, and other components of microtubules and microfilaments. Materials synthesized at
the cell body that have a functional role at the
terminal axon, such as neurotransmitters, are moved
Fig. 7. An electron microscopy (uranyl acetate and lead citrate, 9895) of a peripheral nerve after injury demonstrating the
process of regeneration and degeneration occurring simultaneously in the distal stump. (Right) A classic regenerating unit
containing myelinated and unmyelinated fibers surrounded by perineurium. (Left) The unit demonstrates evidence of Wallerian
degeneration with myelin debris. (From Mackinnon SE, Dellon AL. Surgery of the peripheral nerve. New York: Thieme, 1988.
p. 18; with permission.)
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
at a rapid antegrade rate of up to 410 mm/d. Retrograde transport remains constant at 240 mm/d. It
allows the return of spent neurotransmitter vesicles
and neurotrophic factors for recycling [21,26 – 29].
Basic response to nerve injury
In 1850, Waller [30] described changes in the
distal segment of the hypoglossal nerve of the frog
after transection; these changes later became known
as Wallerian degeneration. His microscopic examination of the changes that occur to the cell body and
axon after nerve transection demonstrated evidence
of distal degeneration. He also described the generation of neural tissue from the end of the proximal
stump, which was later referred to as the ‘‘outgrowth’’ theory. Further investigation by Ramon y
Cajal [31] established that nerve regeneration ema-
115
nated as an outgrowth from the proximal stump of the
severed axon (Fig. 7).
The transection of a peripheral nerve initiates a
cascade of events (Fig. 8). The changes occur in the
cell body, at the site of injury, and in the proximal and
distal segments of the axon. After a central axonotomy, the cell nucleus migrates toward the periphery
of the cell body, and neuronal chromatolysis occurs.
An increase in cellular volume occurs as the cell
increases the production of RNA and associated
regenerative enzymes. The cellular RNA production
and overall metabolism increases at 4 days after
injury, with a peak at 20 days [23,32].
The severity of injury and its proximity to the cell
body influence the degree of neural cell death. The
more proximal the injury is to the cell body, the more
neuronal cell damage occurs. Ygge et al [33] studied
the transection of the rat sciatic nerve and found a
Fig. 8. Schematic diagram of the neural response to axotomy. (A) The normal-appearing cell body of the neuron after a
conceptual scalpel cut in a single axon. Muscle tissue of the end organ also appears normal. (B) Chromatolysis occurs from
nuclear eccentricity, nucleolar swelling, and dissolution of Nissl bodies. The process of Wallerian degeneration begins in the
axon distal to the injury as the denervated muscle begins to atrophy. (C) Histologic evidence of chromatolysis remains in
the neuron as the axon forms a growth cone. Further atrophy of the end-organ muscle occurs. (D) The histologic appearance of
the cell body returns to normal over time. Remaining fibers are pruned down to one as the partially fibrosed muscular end organ
becomes successfully reinnervated. (From Winagrad J, Mackinnon S. Peripheral nerve injuries: repair and reconstruction.
In: Mathis S, Hentz VR, editors. Plastic surgery. Philadelphia: Elsevier and Saunders; 2003; with permission.)
116
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
27% neuronal loss after a proximal injury versus a
7% loss after a distal nerve injury. Similarly, they
found that the more severe the injury, the more
proximal the extent of injury.
Neuronal survival after axonal injury is age
related. It is generally appreciated that younger children have superior functional recovery than adults. In
a 15-year clinical study, Barrios and de Pablos [34]
found that children recovered better than adults after
peripheral nerve injury. However, there is evidence
that a susceptibility to neuronal apoptosis exists in the
neonatal period [35]. In a sciatic nerve transection
model, when comparing a 6- versus 22-day-old animals, significantly greater motoneuron death was
found in the younger group. It has been shown that
the retrograde release of trophic factors from the target
tissue is important for the survival of embryonic
motoneurons [36]. Motoneurons early in the neonatal
period are particularly dependent on growth factors
[35]. Administration of nerve growth factor, ciliary
neurotrophic factor, and neurotrophin-4/5 rescued a
significant number (up to 30%) of motoneurons after
sciatic nerve axotomy in 2- to 5-day-old mice. The
administration of insulin-like growth factor, brainderived neurotrophic factor, and neurotropin-3 rescued the majority of axotomized motoneurons [37].
After injury, Schwann cells proliferate and migrate
as Schwann cell columns (‘‘bands of Bungner’’). The
Schwann cell alters its phenotype and becomes nonmyelinating, with stimulation transforming from a
quiescent state to one of high mitotic activity and
growth factor production [38]. Basement membrane
components and cellular adhesion molecule production increases. This activity of the Schwann cells
enables peripheral growth cone elongation. When
Schwann cell migration is blocked by cytotoxins,
regeneration does not occur [39,40].
Site of injury
Within 24 hours at the site of a peripheral nerve
injury, a single axon produces multiple axonal
sprouts, forming a regenerating unit [41] (Fig. 9).
At the tip of these sprouts is a growth cone that has an
affinity for the fibronectin and laminin of the
Schwann cell basal lamina. Via contact guidance,
filopodia in the growth cone explore the distal environment for the appropriate physical substrate
[42 – 44]. The regeneration proceeds along the neural
tube to reconstitute the axon at a rate between 1 to
4 mm per day. Without successful elongation, a
neuroma forms. With successful reinnervation of the
end organ, the number of axonal sprouts is modulated
over a period of months to years [45].
Distal to the axonotomy site, the nerve segment
undergoes Wallerian degeneration. After transection,
Schwann cell proliferation and myelin breakdown
plays a prominent role. The Schwann cells and
macrophages dispose of the axonal debris that forms
with degeneration. Although the endoneurium and
basement membrane remain essentially intact, the
neural tube that forms eventually collapses as the
myelin and axonal contents are digested. The process
continues until the axons’ neural contents are fully
resorbed, at which point the neural tube becomes
replaced by Schwann cells and macrophages. The
proliferative cells then organize into columns and
form the ‘‘bands of Bungner’’ [46].
Although histologically similar to the Wallerian
degeneration, axonal degeneration is more limited in
the proximal segment. Depending on the severity of
injury, the impact is often limited to one or a few of
the nodes of Ranvier along the proximal segment
[47,48]. The changes that occur in the proximal
segment are referred to as ‘‘traumatic degeneration.’’
In a sufficiently severe injury, in terms of energy and
proximity, the degeneration can result in death of the
cell body.
Contact guidance and neurotropism
In 1905, Ramon y Cajal [31] popularized the
notion of chemical agents ‘‘attracting’’ nerve sprouts
from the proximal stump of a peripheral nerve after
injury. This concept was refined and became known
as neurotropism, or directional guidance, by chemotactic factors. In 1944, Weiss and Taylor [49] demonstrated that neuroregeneration across short nerve
gaps did not seem to be related to an attraction by the
distal stump, the proximal stump, or any other neurotropic effect. They proposed a mechanism of regeneration based on contact guidance. In a series of
Y-shaped artery experiments, they showed that
‘‘. . .in no cases were the regenerating nerves attracted
towards the peripheral nerve fragment but the fiber
stream divided itself more or less evenly regardless of
the kind of destination awaiting them’’ [49]. However, these results were influenced by an immunologic reaction to the aortic allografts (Fig. 10).
Additional work in the ensuing decades has
expanded our understanding of the directional guidance of the regenerating axons. It is believed that that
a motor axon sprouts into a regenerating unit that
contains many axons. After axonotomy, the pioneer
motor axons initially regenerate along motor or sensory Schwann cell tubes. Upon the recognition of a
motor cell tube, the leading motor axon signals back
to the cell body. This results in further communication
Fig. 9. Light microscopy (Toluidine blue, 655) demonstrating a nerve at different levels of a peripheral nerve injury. (A) Wellmyelinated fibers in the proximal aspect. (B) Distal to the nerve injury, Wallerian degeneration is noted with areas of myelin
debris. (C) Each nerve fiber will eventually sprout into a regeneration unit.
118
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
Fig. 10. Schematic diagrams representing four theories of mechanisms of neuronal regeneration. (A) Operative alignment:
Microsurgical technique is used to reapproximate severed nerve endings, thus determining direction. (B) Contact guidance:
Recognition of the appropriate microenvironment guides the regenerating neuron toward the correct distal stump.
(C) Neurotropism: Guidance of the regenerating axon toward the correct distal nerve stump is accomplished by a gradient of
dissolved substances at the molecular level. (D) Neurotropism: Trophic support is supplied to regenerating axons that connect
with appropriate distal nerve stumps. Axons that incorrectly connect undergo degeneration. (From Winagrad J, Mackinnon S.
Peripheral nerve injuries: repair and reconstruction. In: Mathis S, Hentz VR, editors. Plastic surgery. Philadelphia: Elsevier and
Saunders; 2003; with permission.)
Fig. 11. Schematic diagram representing the concept of
preferential motor regeneration after an axonotomy down a
sensory or motor Schwann cell tube. (A) After the nerve is
transected, the nerve fibers sprout into regenerating units.
(B) The motor axons that regenerate into inappropriate
sensory cell tubes do not preferentially mature. (C) The
motor axons, however, are not specifically pruned away.
to the other motor axons within the regenerating unit
to preferentially regenerate within that specific motor
Schwann cell tube. Once the motor axons arrive at the
distal muscle target, the motor axons mature, and
those that regenerate toward sensory targets do not
preferentially mature. However, motor axons that
regenerate into inappropriate Schwann cell tubes are
not specifically ‘‘pruned’’ away, as was previously
thought. This phenomenon is known as ‘‘preferential
motor regeneration’’ [40,50] (Fig. 11).
There is evidence to support the concepts of
contact guidance and neurotropism. Cellular adhesion
molecules (CAMs) have been found to be upregulated in the distal nerve end after axotomy by the
Schwann cell. They include glycoprotein L1, neural
cell adhesion molecule (N-CAM), and N-cadherin
[51]. The expression of L1 and N-CAM is a consequence of the Schwann cells transition from a
myelinated to an unmyelinated state [52]. Williams
[11] has affirmed the significance of an organized
longitudinal fibrin matrix in a model of nerve regeneration. He demonstrated that axons elongate more
effectively along a longitudinally oriented neural
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
matrix than through a randomly oriented fibrin matrix
or in a matrix free chamber. This further supported
the role of the extracellular environment as a scaffold
for the advancing regenerating nerve units.
The term ‘‘neurotropism’’ implies an ability to
stimulate nerve maturation. Neurotrophins that participate in neural regeneration include nerve growth
factor (NGF), brain-derived neurotrophic factor,
neurotrophin 3, neurotropin-4/5, epidermal growth
factor, insulin-like growth factor I and II, and glialderived neurotrophic factor [51,53]. These factors are
primarily produced by Schwann cells, which are
known to become activated in the response to neural
injury. Neurotrophins bind to specialized high-affinity tyrosine kinase receptors and to p75, a low-affinity
NGF receptor [53]. Their role in neural regeneration
is by indirectly acting on the Schwann cells by
stimulating migration and adhesion to axonal projections. Through this mechanism, they play a key role
in the progression of the axonal growth cone. Despite
all these intricate mechanisms that are initiated,
retrograde labeling indicates that up to 51% of the
nerve fibers form inappropriate connections after
significant nerve injury [54].
Nerve repair (end-to-side)
End-to-side neurorrhaphy, or latero-terminal neurorrhaphy, is the method of nerve repair whereby the
distal end of a severed nerve is attached to the side of
an uninjured nerve. In 1903, Balance [55] reported the
first use of the end-to-side neurorrhaphy for the
treatment of a facial palsy by the suturing of the distal
end of the facial nerve laterally into the spinal accessory nerve. This concept was reintroduced in 1991
by Viterbo et al in animal and human studies [56].
Viterbo et al demonstrated in a rat study evidence of
axonal ingrowth and the conduction of electrical
impulses from a healthy nerve into the distal segment
of a severed nerve after an end-to-side repair. Histologic evaluation distal to the repair site showed functional axonal regrowth and muscular reinnervation.
Our understanding of the regeneration process
after an end-to-side neurorrhaphy is evolving. Uninjured sensory axons sprout de novo; however,
whether or not motor axons have the same ability
without an initiating local injury is debated [57].
Tarasidis et al [58] demonstrated an inferior recovery
of motor nerve function when comparing end-to-side
versus the end-to-end technique between posterior
tibial and peroneal nerves in the rat at 16 weeks.
Retrograde labeling showed that the vast majority of
fibers regenerating from the proximal segment of the
nerve were sensory. Noah et al [59] recommended the
119
severance of the perineurium, or an ‘‘epineurial
window,’’ for improved axonal regeneration when
performing an end-to-side repair. They found the
highest axonal counts when a perineurial window or
partial neurectomy was performed [59]. Dvali and
Hunter have recently demonstrated that motor axons
sprout across an end-to-side repair if the donor nerve
is crushed above the repair site, which suggests a
different mechanism for sensory versus motor sprouting (unpublished observation).
Distal receptors
Once denervated, the muscle fibers undergo a
process of atrophy. In 1944, Bowden and Guttmann
[60] performed 140 muscle biopsies on Seddons
patients and identified progressive atrophy and fibrosis
in the first 3 years after denervation. Mammalian
striated muscle after denervation has been shown lose
80% to 90% of its cross-sectional area with no loss of
fiber numbers [5,61]. Additional studies indicated this
process begins within 1 week of denervation [62]. In
cross-sectional analysis, the normally eccentric nucleus of the muscle cell migrates centrally, creating
the ‘‘target cells’’ found in denervated muscle [63,64].
Within 3 months of denervation, particularly if this is
accompanied by lack of activity or movement, interstitial fibrosis replaces the muscle. By 2 to 3 years, the
muscle seems to be essentially replaced by scar tissue
or fat [63].
Reinnervation at the neuromuscular junction is
dependent on the timely arrival of axonal regeneration in the region of the neuromuscular junction. The
absolute time of denervation in which reinnervation is
not possible is not known, but it is likely to occur
after a 12-month period. Experimental evidence
shows that an extended period from injury to repair
further worsens the degree of ultimate recovery [65].
Age also plays a profound role in the ability for
muscles to recover after denervation. Age-related
decreases in muscle mass, force, power, and endurance are in part due to selective denervation of fasttwitch fibers. The reinnervation of denervated muscle
fibers in old animals is not as successful as that seen
in young animals [66 – 68].
Sensory organs
The end organ for sensory fibers consists of one of
variety of specialized structures: Pacinian or Meissner’s corpuscles, Merkel cells, or free nerve endings
(Fig. 12). Pacinian corpuscles receive a single axon
terminal and are less likely to be reinnervated [69,70].
Alternatively, other mechanoreceptors, such as the
120
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
A first-degree peripheral nerve injury (neuropraxia) shows recovery within the first 3 months. A
second-degree injury (axonotmesis) fully recovers;
this occurs at the rate of 1 inch per month. A thirddegree injury demonstrates incomplete recovery with
the same time course as an axonotmetic type injury.
A fourth-degree injury is in continuity, but, like the
transected fifth-degree injury, it will not recover. A
sixth-degree injury demonstrates a variety of recovery patterns based on the pattern and involvement of
the injury.
First-degree injury, neurapraxia
The fundamental basis of a first-degree injury is
a conduction block with preservation of anatomical
Fig. 12. Distal glabrous human skin. MC, Meissner
corpuscle; MD, Merkel cell-neurite complex; PC, Pacinian
corpuscle; SG, sweat gland; SD, sweat duct. (Modified from
Mackinnon SE, Dellon AL. Surgery of the peripheral nerve.
New York: Thieme, 1988. p. 26; with permission.)
Meissner’s corpuscles, Merkel cells, or Ruffini endings, have demonstrated a better ability to be reinnervated [69]. Unlike with muscle injury, sensibility may
be recovered many years after a nerve injury [71].
Clinical classification of nerve injury
In 1941, Cohen introduced a classification to
describe the injury to peripheral nerves: neurapraxia,
axonotmesis, and neurotmesis. In the early 1940s,
Seddon [72,73] examined 650 patients with peripheral nerve injuries and popularized Cohen’s classification system. In 1951, Sunderland [74] expanded
upon the classification system by defining five distinct degrees of nerve injury; this is the classification
system that is commonly used today. Sunderland’s
third- and fourth-degree injuries were included as
extensions of axonotmesis and neurotmesis, respectively, in the original classification used by Seddon.
In 1988, Mackinnon coined the term ‘‘6th degree
injury,’’ which was defined as a mixed injury involving different combinations of injuries [75] (Fig. 13).
Fig. 13. Schematic diagram representing the six classes of
nerve injury. A healthy fascicle is represented where the
blood supply of the nerve enters via the mesoneurium.
Continuing clockwise, Neurapraxia (I) or loss of axonal
myelin is shown. Axonotmesis (II) or second-degree injury
is depicted in the next fascicle where loss of myelin and
axon has occurred without damage to the endoneurium. The
two fascicles (III) depict a third-degree injury. A fourthdegree injury (IV) is shown with damage to all portions of
the nerve except for the epineurium. Fifth-degree injury or
neurotmesis (V) is defined as complete nerve transection.
No recovery is possible without microsurgical repair. A
mixed nerve injury, or sixth-degree injury, is a combination
of various degrees of nerve injury and is referred to as
neuroma in-continuity. (From Winagrad J, Mackinnon S.
Peripheral nerve injuries: repair and reconstruction. In:
Mathis S, Hentz VR, editors. Plastic surgery. Philadelphia:
Elsevier and Saunders; 2003; with permission.)
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
continuity. Although recovery is complete, the time
required varies from days to 3 months. There is no
nerve regeneration involved in the recovery, and
there is no advancing Tinel sign because there is
no axonal involvement. Histologically, at most, demyelination is noted. Examples of first-degree injuries include tourniquet palsy or a similar type of
localized pressure palsy. In early nerve entrapments,
the degree of injury is a first-degree injury with a
partial electrical conduction block with prolonged latency measurements.
Second-degree injury, axonotmesis
In second-degree injuries, the endoneurium and
perineurium remain intact. A Tinel sign is present and
can be traced distally as axonal regeneration occurs.
Disintegration of the axon and all the associated
changes, collectively referred to as Wallerian degeneration, occur below the site of the injury. The general
arrangement of the axon sheaths and the remaining
structures comprising the nerve are preserved, as is
the integrity of the endoneural tube. Because nerve
recovery depends on the regrowth of the axon,
structures typically recover in the anatomic order,
and the preservation of the endoneural tubes ensures
complete restoration of the original pattern of innervation. Nerve recovery should be complete.
Third-degree injury
The structure pattern of third-degree injury
involves endoneurial scarring and disorganization
within the fascicles. The endoneural tube is disrupted,
resulting in erroneous alignment of the regenerating
axonal fibers. Regeneration progresses through a
component of scar tissue within the endoneurium,
thus limiting the regenerating fibers’ ability to make
contact with distal receptors or end organs.
The rate of recovery after a third-degree injury
occurs as expected in axonotmesis, at approximately
1 inch per month. An advancing Tinel sign demonstrates the level of regeneration, but the degree of
recovery will not be complete. If the fascicle contains
sensory and motor fibers, then a mismatch of motor
and sensory fibers can occur, leading to a poorer
functional result.
Fourth-degree injury
A fourth-degree injury is a condition in which the
nerve is physically in continuity, but regeneration
does not occur across scar block. A Tinel sign is
found at the level of the injury but does not advance
121
because regeneration is blocked by scar tissue. As in
second- and third-degree injuries, Wallarian degeneration takes place in the nerve distal to the injury. If
recovery is to occur, then surgical intervention with
nerve repair, nerve grafting, or conduit technique is
necessary. These injuries are typically the result of a
severe stretch, traction, crush, cautery injury, or
nerve injection.
Fifth-degree injury, neurotmesis
The fundamental basis of a fifth-degree injury is
severance of the nerve trunk. Recovery is not possible
without a microsurgical repair. Although these injuries can occur through a severe stretch and resulting
avulsion, they are more commonly associated with
penetrating trauma.
Sixth-degree injury
In 1988, Mackinnon used the term ‘‘sixth-degree
injury, neuroma-in-continuity’’ to describe a mixed
nerve injury [75]. She illustrated how the pattern of
recovery is mixed respective to the varying degree of
injury (I, II, III, IV, V). Partial recovery does occur in
fascicles with a third-degree injury, and complete
recovery takes place in the fascicles that have a firstor second-degree injury. No recovery is seen in
fascicles with fourth- or fifth-degree injury patterns.
In this injury, some fascicles may be normal.
The sixth-degree injury is in many ways the most
surgically challenging. There is a significant potential
for underestimating the overall injury due to limited
recovery in the less injured portions of the nerve. A
careful assessment and correlation of the observed
recovery pattern and fascicular anatomy is indicated.
Similarly, careful surgical technique with microinternal neurolysis allows for fascicular separation and
individual reconstruction of the fourth- and fifthdegree injuries without injury to the other fascicles.
Compression neuropathy
The pathology found in chronic nerve compression can span the entire range of the nerve injury. In
the early stages, nerve compression may be associated with the breakdown in the blood-nerve barrier. It
is followed by subperineurial edema with fibrosis and
subsequently localized segmental demyelination. Diffuse demyelination and eventually Wallerian degeneration of the nerve fibers occurs in advanced stages.
These changes are dependent on the amount and
extent of the compressive forces (Fig. 14).
122
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
Fig. 14. Schematic diagram outlining the pathogenesis of nerve compression. The initial changes involve the blood-nerve barrier
with endoneurial and subperineurial edema secondary to subperineurial edema. Next, changes involve the connective tissue
layers that demonstrate increased perineurial and epineurial thickening. Localized nerve fiber changes occur, with some fascicles
appearing normal and others demonstrating localized areas of demyelination of the large fibers. Unmyelinated fibers show
evidence of regeneration with a new population of small unmyelinated fibers. The central fascicles or the central fibers are
usually spared, and progressive compression results in diffuse injury of the myelinated and unmyelinated fibers. (Adapted from
Mackinnon SE, Dellon AL. Surgery of the peripheral nerve. New York: Thieme, 1988. p. 42; with permission.)
Nerves with more connective tissue and fewer
fascicles may be better protected and undergo neural
changes slower than nerves with less connective
tissue [76]. Similarly, fascicles located closer to the
predominant compression undergo changes sooner
than other more distantly located fascicles. Organized
fibrosis in the subperineurial space results in the
formation of Renault bodies that often occur at sites
where the nerve crosses a joint. This is believed to be
related to repetitive movement and traction occurring
across joints.
Various methods have been described to recreate
or induce acute or chronic compression in a peripheral nerve. Rydevik et al [77] used a rabbit tibial
nerve to examine the effect of graded compression on
the intraneural blood flow. They found that external
pressures of 20 mm Hg caused a reduction in venule
blood flow, 30 mm Hg resulted in an inhibition of
axonal transport, and 80 mm Hg caused a complete
cessation of intraneural blood flow [77].
The impact of prolonged nerve compression on
function has also been examined. In a rat model, the
effects of various pressures (10, 30, and 80 mm Hg)
over different durations (4 hours to 28 days) were
studied [78]. Within hours, subperineurial edema,
inflammation, and the formation of fibrin deposits
were reported; within days, fibrous tissue proliferation was present; and fibrosis was seen by 28 days.
At higher pressures, (80 mm Hg), axonal damage
was evident.
In 1973, Upton and McComas [79] introduced the
‘‘double crush’’ hypothesis. They hypothesized that
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
123
proximally located nerve compression could cause
distal sites to become susceptible to compression.
The authors associated a high incidence of carpal
and cubital tunnel syndromes with associated cervical
root lesions. Summation of compression along the
nerve can lead to changes in axoplasmic flow with
subsequent pathology. The concept of double or multiple crush has a clinical relevance in patients who
demonstrate multiple levels of nerve compression.
Failure to recognize this possible condition results in
a failure to relieve the patient’s underlying pathology.
Systemic disease, such as diabetes, thyroid disease, alcoholism, and various arthritic states, can
result in peripheral neuropathies and lead to enhanced
susceptibility of nerve compression. Dellon et al [80]
have demonstrated that, in the streptozocin model, the
diabetic nerve is more susceptible to compression. It
seems that glucose enters the nerve directly (because
there is no blood-nerve barrier to glucose), causing
endoneurial edema. The glucose is then metabolized
to hydrophilic polyols, which pull additional water
into the endoneurial space. The resulting edema leads
to increased endoneurial fluid pressure and progresses to myelin changes. It seems that anything that
could alter axoplasmic physiology could render the
nerve more susceptible to developing a compression
neuropathy and act as a ‘‘crush.’’
Vibration can also result in peripheral neuropathy.
A number of studies have shown the association of
vibratory exposure with the development of peripheral neuropathy [81 – 84]. The histologic changes
associated with this type of exposure are much like
those seen in compression neuropathy, with the
development of intraneural edema, demyelination,
and eventual axonal loss [85]. Epidemiologic investigations have found a strong association between
vibratory exposure and the development of compression at the carpal tunnel [86].
electrical stimulation of the decompressed nerve is
beneficial by resulting in a more synchronous regeneration. Nicolaidis and Williams [92] presented
some promising work on the usefulness of continuous electrical stimulation via implantable electrodes
in deinnervated muscle. In a study of 15 patients
with peripheral nerve injuries of the upper extremities, the authors demonstrated improved functionality in all cases with extended stimulation (range of
127 to 346 days).
Improvements in neuroregeneration
References
Recently, an enhancement in neuroregeneration
with the use of FK506 has been shown after nerve
crush, transection, repair, and nerve grafting. Although the exact mechanism of this action is unclear,
it seems that its beneficial effects are enhanced at
higher doses or with immunosuppressive doses
[87 – 90]. It also seems that the beneficial effects of
FK506 are seen even when its administration is
delayed [91].
Al-Majed et al [50] have recently shown the
application of direct nerve stimulation to increase
peripheral never regeneration. It is believed that
Summary
The response to nerve injury is a complex and
often poorly understood mechanism. An in-depth and
current command of the relevant neuroanatomy, classifications systems, and responses to injury and
regeneration are critical to current clinical success.
Continued progress must be made in our current
understanding of these varied physiologic mechanisms of neuro-regeneration if any significant progress in clinical treatments or outcome is to be
expected in the future. Reconstructive surgeons have
in many ways maximized the technical aspects of
peripheral nerve repair. However, advances in functional recovery may be seen with improvements in
sensory and motor rehabilitation after peripheral
nerve surgery and with a combined understanding
of the neurobiology and neurophysiology of nerve
injury and regeneration.
Acknowledgments
We thank DA Hunter, research assistant, Washington University School of Medicine, St. Louis,
for his contribution to the electron and light microscopy images.
[1] Key A, Retzius G. Studien in der anatomie des nervensystems und des binfegwebes. Stockholm: Samson
and Wallin; 1876.
[2] Millesi H. The nerve gap: theory and clinical practice.
Hand Clin 1986;2:651 – 63.
[3] Millesi H, Zoch G, Rath T. The gliding apparatus of
peripheral nerve and its clinical significance. Ann Chir
Main Memb Super 1990;9:87 – 97.
[4] Wilgis EF, Murphy R. The significance of longitudinal
excursion in peripheral nerves. Hand Clin 1986;2:
761 – 6.
[5] Sunderland S. Nerves and nerve injuries. Edinburgh:
Churchill Livingstone; 1978.
124
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
[6] Reale E, Luciano L, Spitznas M. Freeze-fracture faces
of the perineurial sheath of the rabbit sciatic nerve.
J Neurocytol 1975;4:261 – 70.
[7] Olsson Y, Reese TS. Permeability of vasa nervorum
and perineurium in mouse sciatic nerve studied by
fluorescence and electron microscopy. J Neuropathol
Exp Neurol 1971;30:105 – 19.
[8] Sunderland S. The anatomy and physiology of nerve
injury. Muscle Nerve 1990;13:771 – 84.
[9] Sunderland S. The intraneural topography of the radial,
median and ulnar nerves. Brain 1945;68:243 – 99.
[10] Jabaley ME, Wallace WH, Heckler FR. Internal topography of major nerves of the forearm and hand:
a current view. J Hand Surg [Am] 1980;5:1 – 18.
[11] Williams HB, Jabaley ME. The importance of internal
anatomy of the peripheral nerves to nerve repair in the
forearm and hand. Hand Clin 1986;2:689 – 707.
[12] Breidenbach WC, Terzis JK. The blood supply of vascularized nerve grafts. J Reconstr Microsurg 1986;
3:43 – 58.
[13] Lundborg G. Structure and function of the intraneural
microvessels as related to trauma, edema formation,
and nerve function. J Bone Joint Surg Am 1975;57:
938 – 48.
[14] Ogata K, Manske PR, Lesker PA. The effect of surgical
dissection on regional blood flow to the ulnar nerve in
the cubital tunnel. Clin Orthop 1985;193:195 – 8.
[15] Ogata K, Naito M. Blood flow of peripheral nerve
effects of dissection, stretching and compression.
J Hand Surg [Br] 1986;11:10 – 4.
[16] Smith CE, Atchabahian A, Mackinnon SE, et al. Development of the blood-nerve barrier in neonatal rats.
Microsurgery 2001;21:290 – 7.
[17] Lundborg G, Dahlin LB. Anatomy, function, and
pathophysiology of peripheral nerves and nerve compression. Hand Clin 1996;12:185 – 93.
[18] Vizoso A. The relationship between internodal length
and growth in human nerves. J Anat 1984;84:342.
[19] Winagrad J, Mackinnon S. Peripheral nerve injuries:
repair and reconstruction. In: Mathis S, Hentz VR,
editors. Plastic surgery. Philadelphia: Elsevier and Saunders; 2003. in press.
[20] Watchmaker G, Mackinnon SE. Nerve injury and repair. In: Peimer CA, editor. Surgery of the hand and
upper extremity. New York: McGraw-Hill; 1996.
p. 1251 – 76.
[21] Dvali LT, Mackinnon SE. Basic pathophysiology of
the hand, wrist, and forearm. In: Weiss AC, Berger
RA, editors. Hand surgery. Philadelphia: Lippincott
Williams and Wilkins; 2003. in press.
[22] Ducker TB, Kempe LG, Hayes GJ. The metabolic
background for peripheral nerve surgery. J Neurosurg
1969;30:270 – 80.
[23] Ducker TB, Kauffman FC. Metabolic factors in surgery of peripheral nerves. Clin Neurosurg 1977;24:
406 – 24.
[24] Bodian D, Howe H. Experimental studies of intraneural spead of poliomyelitis virus. Bull Johns Hopkins
Hosp 1941;68:248 – 67.
[25] Droz B, Leblond C. Migration of proteins along the
axons of the sciatic nerve. Science 1962;137:1047.
[26] Droz B, Rambourg A, Koenig HL. The smooth endoplasmic reticulum: structure and role in the renewal of
axonal membrane and synaptic vesicles by fast axonal
transport. Brain Res 1975;93:1 – 13.
[27] Lasek J, Shelanski M, Brinkley B, et al. Cytoskeletons
and the architecture of nervous systems. Neurosci Res
Program Bull 1981;19:1 – 153.
[28] Lubinska L, Niemierko S. Velocity and intensity of
bidirectional migration of acetylcholinesterase in transected nerves. Brain Res 1971;27:329 – 42.
[29] Stockel K, Schwab M, Thoenen H. Comparison between the retrograde axonal transport of nerve growth
factor and tetanus toxin in motor, sensory and adrenergic neurons. Brain Res 1975;99:1 – 16.
[30] Waller A. Experiments on the glossopharyngeal and
hypoglossal nerves of the frog and observations produced thereby in the structure of their primitive fibers.
Phil Trans R Soc London 1850;140:423.
[31] Ramon y Cajal S. Mechanismo de la degeneration y
regeneration de nervos. Trab Lab Inbest Biol 1905;
9:119.
[32] Edstrom A, Hanson M. Retrograde axonal transport of
proteins in vitro in frog sciatic nerves. Brain Res 1973;
61:311 – 20.
[33] Ygge J. Neuronal loss in lumbar dorsal root ganglia
after proximal compared to distal sciatic nerve resection: a quantitative study in the rat. Brain Res 1989;
478:193 – 5.
[34] Barrios C, de Pablos J. Surgical management of nerve
injuries of the upper extremity in children: a 15-year
survey. J Pediatr Orthop 1991;11:641 – 5.
[35] Lowrie MB, Vrbova G. Dependence of postnatal motoneurones on their targets: review and hypothesis.
Trends Neurosci 1992;15:80 – 4.
[36] Kuno M. Target dependence of motoneuronal survival:
the current status. Neurosci Res 1990;9:155 – 72.
[37] Li L, Oppenheim RW, Lei M, et al. Neurotrophic
agents prevent motoneuron death following sciatic
nerve section in the neonatal mouse. J Neurobiol
1994;25:759 – 66.
[38] Curtis R, Stewart HJ, Hall SM, et al. GAP-43 is expressed by nonmyelin-forming Schwann cells of the
peripheral nervous system 2. J Cell Biol 1992;116:
1455 – 64.
[39] Brushart TM, Seiler WA. Selective reinnervation of
distal motor stumps by peripheral motor axons. Exp
Neurol 1987;97:289 – 300.
[40] Brushart TM, Gerber J, Kessens P, et al. Contributions
of pathway and neuron to preferential motor reinnervation. J Neurosci 1998;18:8674 – 81.
[41] Morris JH, Hudson AR, Weddell G. A study of degeneration and regeneration in the divided rat sciatic nerve
based on electron microscopy. II. The development of
the ‘‘regenerating unit’’. Z Zellforsch Mikrosk Anat
1972;124:103 – 30.
[42] Letourneau PC, Wessells NK. Migratory cell locomotion versus nerve axon elongation: differences based
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
on the effects of lanthanum ion. J Cell Biol 1974;61:
56 – 69.
Letourneau PC. Cell-to-substratum adhesion and guidance of axonal elongation. Dev Biol 1975;44:92 – 101.
Letourneau PC. Chemotactic response of nerve fiber
elongation to nerve growth factor. Dev Biol 1978;66:
183 – 96.
Mackinnon SE, Dellon AL, O’Brien JP. Changes in
nerve fiber numbers distal to a nerve repair in the rat
sciatic nerve model. Muscle Nerve 1991;14:1116 – 22.
Fawcett JW, Keynes RJ. Peripheral nerve regeneration.
Annu Rev Neurosci 1990;13:43 – 60.
Horch K. Central responses of cutaneous sensory neurons to peripheral nerve crush in the cat. Brain Res
1978;151:581 – 6.
Kristensson K. Retrograde signalling of nerve cell
body response to trauma. In: Glorio A, Millesi H,
Mingrino S, editors. Posttraumatic peripheral nerve
regeneration. New York: Raven Press; 1981. p. 27 – 34.
Weiss P, Taylor A. Further experimental evidence
against neurotropism in nerve. J Exp Zool 1944;
95:233 – 57.
Al-Majed AA, Neumann CM, Brushart TM, et al. Brief
electrical stimulation promotes the speed and accuracy
of motor axonal regeneration. J Neurosci 2000;20:
2602 – 8.
Fu SY, Gordon T. The cellular and molecular basis of
peripheral nerve regeneration. Mol Neurobiol 1997;
14:67 – 116.
Martini R. Expression and functional roles of neural
cell surface molecules and extracellular matrix components during development and regeneration of peripheral nerves. J Neurocytol 1994;23:1 – 28.
Mendell LM. Neurotrophic factors and the specification of neural function. Neuroscientist 1995;1:26 – 34.
Shinohara H, Naora H, Hashimoto R, et al. Development of the innervation pattern in the upper limb of
staged human embryos. Acta Anat (Basel) 1990;138:
265 – 9.
Ballance C, Ballance HSP. Remarks on the operative
treatment of chronic facial palsy of peripheral origin.
BMJ 1903;2:1009 – 13.
Viterbo F, Trindade JC, Hoshino K, et al. Latero-terminal neurorrhaphy without removal of the epineural
sheath: experimental study in rats. Rev Paul Med 1992;
110:267 – 75.
Yan JG, Matloub HS, Sanger JR, et al. A modified
end-to-side method for peripheral nerve repair: large
epineurial window helicoid technique versus small epineurial window standard end-to-side technique. J Hand
Surg [Am] 2002;27:484 – 92.
Tarasidis G, Watanabe O, Mackinnon SE, et al. End-toside neurorrhaphy resulting in limited sensory axonal
regeneration in a rat model. Ann Otol Rhinol Laryngol
1997;106:506 – 12.
Noah EM, Williams A, Jorgenson C, et al. End-to-side
neurorrhaphy: a histologic and morphometric study of
axonal sprouting into an end-to-side nerve graft. J Reconstr Microsurg 1997;13:99 – 106.
125
[60] Bowden R, Guttman E. Denervation and reinnervation
of human voluntary muscle. Brain 1944;67:273.
[61] Sunderland S. Denervation changes in mammalian
striated muscle. J Neurol Neurosurg Psychiatry 1950;
13:159.
[62] Guttmann E. Effect of delay of innervation on recovery
of muscle after nerve lesions. J Neurophysiol 1948;
11:277.
[63] Aird RB, Naffziger HC. The pathology of human striated muscle following denervation. J Neurosurg 1953;
10:216 – 27.
[64] Cancilla PA. General reactions of muscle to injury.
In: Heffner RR, editor. Muscle pathology: contemporary issues in surgical pathology. New York: Churchill
Livingstone; 1984. p. 15 – 30.
[65] Fu SY, Gordon T. Contributing factors to poor functional recovery after delayed nerve repair: prolonged
denervation. J Neurosci 1995;15:3886 – 95.
[66] Brooks SV, Faulkner JA. Skeletal muscle weakness in
old age: underlying mechanisms. Med Sci Sports Exerc
1994;26:432 – 9.
[67] Faulkner JA, Brooks SV, Zerba E. Muscle atrophy and
weakness with aging: contraction-induced injury as an
underlying mechanism. J Gerontol Biol Sci 1995;50A:
124 – 9.
[68] Kobayashi J, Mackinnon SE, Watanabe O, et al. The
effect of duration of muscle denervation on functional
recovery in the rat model. Muscle Nerve 1997;20:
858 – 66.
[69] Krishnamurti A, Kanagasuntheram R, Vij S. Failure of
reinnervation of Pacinian corpuscle after nerve crush:
an electron microscopic study. Acta Neuropathol (Berl)
1973;23:338 – 41.
[70] Mackel R, Kunesch E, Waldhor F, et al. Reinnervation
of mechanoreceptors in the human glabrous skin following peripheral nerve repair. Brain Res 1983;268:
49 – 65.
[71] Novak CB, Kelly L, Mackinnon SE. Sensory recovery
after median nerve grafting. J Hand Surg [Am] 1992;
17:59 – 68.
[72] Seddon H. War injuries of peripheral nerves [abstract].
Br J Surg 1942;2:235.
[73] Seddon H. Three types of nerve injury [abstract]. Brain
1943;66:237.
[74] Sunderland S. A classification of peripheral nerve injuries producing loss of function. Brain 1951;74:491.
[75] Mackinnon SE. New directions in peripheral nerve
surgery. Ann Plast Surg 1989;22:257 – 73.
[76] Lundborg G. Nerve injury and repair. New York:
Churchill Livingstone; 1988.
[77] Rydevik B, Lundborg G, Bagge U. Effects of graded
compression on intraneural blood blow: an in vivo
study on rabbit tibial nerve. J Hand Surg [Am] 1981;
6:3 – 12.
[78] Powell HC, Myers RR. Pathology of experimental
nerve compression. Lab Invest 1986;55:91 – 100.
[79] Upton AR, McComas AJ. The double crush in nerve
entrapment syndromes. Lancet 1973;2:359 – 62.
[80] Dellon AL, Mackinnon SE, Seiler WA. Susceptibility
126
[81]
[82]
[83]
[84]
[85]
[86]
S.P. Maggi et al. / Clin Plastic Surg 30 (2003) 109–126
of the diabetic nerve to chronic compression. Ann Plast
Surg 1988;20:117 – 9.
Ho ST, Yu HS. Ultrastructural changes of the peripheral nerve induced by vibration: an experimental study.
Br J Ind Med 1989;46:157 – 64.
Lundborg G, Dahlin LB, Danielsen N, et al. Intraneural
edema following exposure to vibration. Scand J Work
Environ Health 1987;13:326 – 9.
Lundborg G, Dahlin LB, Hansson HA, et al. Vibration
exposure and peripheral nerve fiber damage. J Hand
Surg [Am] 1990;15:346 – 51.
Stromberg T, Lundborg G, Holmquist B, et al. Impaired
regeneration in rat sciatic nerves exposed to short-term
vibration. J Hand Surg [Br] 1996;21:746 – 9.
Stromberg T, Dahlin LB, Brun A. Structural nerve
changes at wrist level in workers exposed to vibration.
Occup Environ Med 1997;54:307 – 11.
Wieslander G, Norback D, Gothe CJ, et al. Carpal
tunnel syndrome (CTS) and exposure to vibration, repetitive wrist movements, and heavy manual work:
a case-referent study. Br J Ind Med 1989;46:43 – 7.
[87] Doolabh VB, Mackinnon SE. FK506 accelerates functional recovery following nerve grafting in a rat model.
Plast Reconstr Surg 1999;103:1928 – 36.
[88] Hare GM, Mackinnon SE, Midha R, et al. Cyclosporine A inhibits lymphocyte migration into ovine
peripheral nerve allografts. Microsurgery 1996;17:
697 – 705.
[89] Mackinnon SE, Hudson AR, Bain JR, et al. The
peripheral nerve allograft: an assessment of regeneration in the immunosuppressed host. Plast Reconstr
Surg 1987;79:436 – 46.
[90] Yang RK, Lowe IJB, Sobol JB, et al. Dose-dependent
effects of FK506 on neuroregeneration in a rat model.
Plast Reconstr Surg 2003 in press.
[91] Sobol JB, Lowe IJB, Yang RK, et al. Effects of delaying FK506 administration on neuroregeneration in a
rodent model. J Reconstr Microsurg 2003;19:113 – 8.
[92] Nicolaidis SC, Williams HB. Muscle preservation
using an implantable electrical system after nerve injury and repair. Microsurgery 2001;21:241 – 7.