Unilateral Nerve Injury Produces Bilateral

Loss of Distal Innervation

Anne Louise Oaklander, MD, PhD, and Jennifer M. Brown, BS

There are no known anatomical connections between neurons that innervate homologous right and left body parts.
Nevertheless, some patients develop bilateral abnormalities after unilateral injury, a phenomenon often unrecognized and
not yet characterized. Therefore, we examined in rats the effects of ligating and cutting one tibial nerve on sensory
function and on density of innervation in hind paws contralaterally as well as ipsilaterally to the injury, at times between
1 day and 5 months after surgery. Punches removed from tibial- or sural-innervated plantar paw skin were immunola-
beled to quantitate epidermal nerve endings. Naive and sham-operated rats provided controls. Axotomized rats had
near-total loss of PGP9.5ⴙ innervation within ipsilateral tibial-innervated skin at all time-points. Adjacent ipsilateral
sural-innervated skin had persistent hyperalgesia without denervation, and robust axonal sprouting at 5 months after
surgery. Contralesional hind paws lost 54% of innervation in tibial-innervated epidermis starting 1 week after surgery
and persisting throughout. Contralesional sural-innervated skin had neither neurite loss nor sprouting. These results
imply that unilateral nerve injury can cause profound, long lasting, nerve–branch–specific loss of distal innervation
contralaterally as well as ipsilaterally. They discredit the practice of using tissues contralateral to an injury to provide
normative controls and suggest the possibility of rapid, transmedian postinjury signals between homologous mirror-
image neurons.
Ann Neurol 2004;55:639 – 644

Neuroanatomy textbooks show no connections be-
tween neurons that innervate homologous areas of the
left and right sides of the body, and injuries usually are
not thought to directly affect the opposite side. Despite
this, several studies suggest that unilateral injury affects
contralateral (contralesional) tissues as well as those ip-
silateral (ipsilesional) to the injury. Usually, contrale-
sional effects are restricted to specific areas, most com-
monly those that are mirror-image to tissues innervated
by injured nerves. This pattern demonstrates that the
signals initiating contralesional effects are not systemic
and suggests neural mediation.
Early reports of patients with contralesional effects
come from S. Weir Mitchell, whose 1872 masterpiece,
Injuries of Nerves and Their Consequences, presents ex-
aminations of Civil War soldiers with meticulously
characterized nerve and plexus injuries. Several with
single-limb wounds developed bilateral- or contralateral-
only pain or sensory loss (pp. 137, 146).1 Modern
studies confirm contralateral changes in some patients,
including limb edema, loss of strength, and changes in
bone metabolism, suggesting effects beyond mere al-
tered use.2 Examination of skin biopsies PGP9.5-
immunolabeled to show nerve endings has demon-
strated that patients with postherpetic neuralgia (PHN)
after unilateral shingles have lost most innervation in
previously affected skin.3 On average, they have also
lost half of the innervation in clinically asymptomatic
mirror-image contralateral skin.3 These findings, now
confirmed,4 were attributed by many to undetected
spread of Varicella zoster virus into the spinal cord.
However, because we, and others, have found contrale-
sional effects after one-sided experimental nerve inju-
ries where there is no infection,5–7 we considered the
alternative: that contralateral loss of distal neurites is
initiated by ipsilateral axonal injury alone.
Animal studies have provided indirect support.6
Contralesional changes in animals include changes
within sensory ganglia8 and spinal cord, some develop-
ing just minutes after paw injury.9 Distal contralesional
changes include reduced extravasation of intravascular
proteins (a C-fiber axon reflex)10 and formation of ad-
ditional neuromuscular junctions in muscles mirror-
image to those denervated by unilateral axotomy.11 Re-
ports of bilateral hind paw hyperalgesia after unilateral
injury demonstrate that contralesional changes affect
behavior.12,13 Systemic lidocaine has different effects
against the hyperalgesia that develops in the paws ipsi-

From the Nerve Injury Unit, Departments of Anesthesiology, Neu- Address correspondence to Dr Oaklander, Massachusetts General
rology, and Neuropathology, Massachusetts General Hospital, Har- Hospital, 55 Fruit Street, Clinics 3, Boston, MA 02114.
vard Medical School, Boston, MA. E-mail: aoaklander@partners.org
Received Aug 14, 2003, and in revised form Oct 28 and Nov 11,
2003 and Jan 3, 2004. Accepted for publication Jan 3, 2003.
Published online Apr 12, 2004, in Wiley InterScience
(www.interscience.wiley.com). DOI: 10.1002/ana.20048

© 2004 American Neurological Association 639

Published by Wiley-Liss, Inc., through Wiley Subscription Services
lateral and contralateral to nerve ligation.14 Even Aply-
sia californica develops bilateral biochemical and behav-
ioral changes after unilateral nerve crush.15
Nerve injury reduces cutaneous innervation in the
appropriate distal cutaneous territory ipsilaterally16 and
causes later collateral sprouting from adjacent unin-
jured nerves into hypoinnervated areas.17 To investi-
gate its contralesional effects, we initiated Wallerian de-
generation without infection by performing unilateral
axotomies and evaluating the effects on innervation
within skin samples from contralateral as well as ipsile-
sional paw. The results demonstrate profound, long-
lasting contralateral anatomic changes that correspond
precisely to those described in humans.
Materials and Methods
Animal Use and Creation of Lesions
Male Sprague-Dawley rats (200 –250gm; Charles River Labs,
Wilmington, MA; Iffa Credo, L’Arbresle, France) were stud-
ied. Procedures were approved by Animal Care and Use
Committees and conformed to ethical guidelines of the In-
ternational Association for the Study of Pain.18 Groups of
same-age naive rats were purchased, housed, tested, and eu-
thanized among experimental rats. Naive rats were eutha-
nized 1 week (n ⫽ 7), 3 weeks (n ⫽ 11), or 5 months (n ⫽
6) after baseline testing. Neurite densities from left and right
hind paws of naive rats euthanized 3 weeks after baseline
were compared to determine if left-right differences are con-
stitutively present.
Halothane (2%) provided surgical anesthesia. Six rats un-
derwent sham surgery (the left sciatic was exposed only) and
were euthanized 3 weeks later. In experimental rats, the left
common peroneal and tibial branches were ligated with 5.0
silk and transected just distally (spared nerve injury [SNI]19;
Fig 1A). The sural branch was uninjured. Experimental rats
were euthanized 1 day (n ⫽ 6), 3 days (n ⫽ 6), 1 week (n ⫽
7), 3 weeks (n ⫽ 6), and 5 months (n ⫽ 5) after surgery.
Fig 1. (A) Location of sites of axotomy; (B) innervation of rat
plantar paw. Modified and reprinted with permission from
Decosterd and Woolf.19
640 Annals of Neurology Vol 55 No 5 May 2004
Documentation of Hyperalgesia
Two distinct areas of the proximal plantar left and right hind
paw lesion were tested using established protocols.19 The
paw center, between tori, represented tibial-innervated skin;
proximolateral glabrous skin represented sural-innervated
skin (see Fig 1B). All rats were habituated to testing for 1
week and tested in random order by a blinded examiner.
Two consecutive days of testing established baseline func-
tion. For the 3-week time point, testing occurred on days 3,
7, 14, and 21 after the day of surgery. For the 5-month time
point, testing was at 3 weeks, 5 weeks, 3 months, and 5
months after the day of surgery.
Mechanosensation was measured on a wire grid by stim-
ulating the above locations with von Frey monofilaments.
The threshold for paw withdrawal was defined as the thin-
nest monofilament that produced withdrawal on at least one
of five trials.20 Mechanical nociception was tested by timing
by stopwatch the duration of paw withdrawal after prick
with a safety pin. For thermal testing, a drop of acetone was
applied to the plantar paw without touching the skin. The
duration of paw withdrawal was recorded by stopwatch with
a minimum of 0.5 seconds and a cutoff of 20 seconds.21
Acquisition, Processing, and Measurements from Skin
Biopsies
At completion of behavioral testing, rats were euthanized by
halothane overdose, and one 2mm diameter punch was re-
moved from each of the four tested areas. Punches were im-
munolabeled against pan-neuronal marker protein gene
product (PGP) 9.5, a ubiquitin carboxyl-terminal hydrolase
(Chemicon, Temecula, CA)22 that enables quantitation of
individual epidermal axons using standard methods.23 The
neural identity of cutaneous PGP9.5⫹ neurite profiles has
been confirmed ultrastructurally.24,25 Slides were masked and
randomized to conceal their identity until data acquisition
was complete. All PGP9.5-immunolabeled neurites within
the entire thickness of epidermis in four randomly chosen
sections were counted.
Data Analysis
Results are presented as mean ⫾ SEM. Data comparing
groups of rats were analyzed by ANOVA and Scheffé post
hoc or Student’s t tests, assuming unequal variances. SAS 8.1
software (SAS Institute, Cary, NC) was used.
Results
Behavioral Testing
The results of behavioral testing of the lesioned rats
duplicate those previously published.19 Profound, long-
lasting mechanical allodynia and hyperalgesia, as well
as cold hyperalgesia, were present in all rats in the ip-
silateral sural-innervated territory only (Fig 2). Results
from sham-operated rats did not differ from those
from naive rats.
Neurite Densities from Naive and Sham-operated
Control Rats
In naive rats (n ⫽ 12), t test of paired values demon-
strated no difference in neurite densities between
Fig 2. Mechanical allodynia at sural-innervated plantar paw site: error bars represent SEM. (circles) Data from naive rats;
(squares) data from sham-operated rats; (diamonds) data from axotomized rats.

punches from the left (847 ⫾ 47) or right (952 ⫾ 28)
paw ( p ⱕ 0.36); therefore, data from left paws of naive
rats were used as controls. Sham operation in 12 rats
killed 3 weeks after surgery did not reduce the density
of innervation from that present in same-age naive rats
( p ⱖ 0.98), so naive rats provided control values. Neu-
rite densities at the tibial and sural biopsy sites differed
at the 5-month time point only ( p ⱕ 0.03), so data
from tibial- and sural-innervated skin punches from
naive rats were analyzed independently to provide sep-
arate controls for tibial- and sural-innervated samples.
ANOVA and post hoc analysis showed no significant
time effect ( p ⱕ 0.41); therefore, data from naive rats
used for the 7-day postoperative time point were used
as controls against which to compare data from rats
that underwent axotomy 1, 3, or 7 days before death.
Neurite Densities in Skin from the Paw Ipsilateral to
Nerve Injury (Ipsilesional Paw)
Neurite densities from rats that had undergone axo-
tomy were compared with data from same-age naive
rats (see Fig 4) to determine the percentages given be-
low. At 1 and 3 days only, postoperative edema in the
ipsilesional paw artifactually lowered densities in sural
(and presumably in tibial) territories by half, so these
data were discarded. However, the near-total neurite
loss in ipsilateral tibial-innervated punches was far
greater: 97% loss at 1 day ( p ⱕ 0.03) and 99.8% loss

Fig 3. Fluorescein isothiocyanate–conjugated PGP9.5⫹ epidermal and dermal innervation in tibial-innervated plantar paw skin at
3 weeks after surgery. Near-total loss of PGP9.5⫹ neurites is evident ipsilaterally, and loss of half of neurites is visible in contrale-
sional uninjured mirror-image skin. SNI ⫽ spared nerve injury.

Oaklander and Brown: Bilateral WD 641

Fig 4. Densities of PGP9.5⫹ epidermal innervation in naive and axotomized rats at all time points.
at 3 days ( p ⱕ 0.000006). Postoperative paw edema
after sciatic nerve injury lasts 5 days,26 and by 7 days
after surgery, values from ipsilateral sural-innervated
punches had returned to baseline ( p ⱕ 0.37), whereas
near-total neurite loss persisted in tibial-innervated sam-
ples. Results were similar at the 3-week time point, with
loss of more than 99% of neurites in tibial-territory
punches but no reduction of sural innervation ( p ⱕ
0.16; Fig 3). At 5 months after surgery, neurite reduc-
tion in tibial-innervated skin samples had moderated to
85% of control value (128 ⫾ 37 neurites/mm2 skin sur-
face area; p ⱕ 4.1 ⫻10⫺6). Density in sural-innervated
skin had increased to 161% of control (1,551 ⫾ 105
neurites/mm2 skin surface area; p ⱕ 0.01).
Neurite Densities from the Paw Contralateral to
Nerve Injury (Contralesional Paw)
Data from the contralesional paw (Fig 4) were analyzed
identically to data from ipsilesional paws. No postop-
erative paw edema developed contralaterally. Neurite
densities from sural-innervated punches did not differ
from control values at any time point ( p ⱕ 0.85).
Neurite densities in punches from tibial-innervated
skin were reduced to 54.2% of control values at time
points between 1 week and 5 months after surgery
( p ⱕ 0.0001; see Figs 3 and 4).
642 Annals of Neurology Vol 55 No 5 May 2004
Discussion
This study demonstrates long-lasting loss of half of the
epidermal neurites in skin mirror-image-contralateral
to skin innervated by transected tibial branches of the
sciatic nerve. Changes ipsilaterally comprised near-total
loss of innervation in the area of the paw innervated by
the transected nerve, and apparent neurite sprouting in
the territory of the adjacent uninjured nerve. Corre-
sponding loss of dermal axons accompanied epidermal
losses, although the bundling of dermal axons pre-
cludes quantitative analysis.
Ipsilateral losses of distal innervation after axotomy
reflect Wallerian degeneration,25 and there is prelimi-
nary evidence that this may be true contralaterally.
Watson and colleagues performed an autopsy on the
corpse of a woman with unilateral PHN after unilateral
facial zoster and found “obvious pathological involve-
ment of the contralateral ophthalmic and supraorbital
nerves.”4 The damage, less severe than that seen ipsile-
sionally, comprised degeneration of large-diameter fi-
bers, demyelinated axons, and many thinly myelinated
axons thought to indicate earlier axonal degeneration
and regeneration. These authors also reevaluated their
previously published autopsies of patients after truncal
zoster and discovered unappreciated contralateral as
well as ipsilateral axonal loss within nerve trunks and
roots.4 Therefore, there is evidence that contralesional
damage is present proximally as well as distally. These
findings argue against the alternative hypothesis that
the distal changes described here represent focal
changes in axonal properties that reduce PGP9.5 im-
munolabeling without causing axonal degeneration. An
important question is whether distal contralesional
degenerative-like changes also affect central axon termi-
nals of contralateral primary sensory neurons and dis-
connect them from the central nervous system.
Our study demonstrates that ipsilateral hyperalgesia
in sural-innervated skin was accompanied by normal or
increased densities of epidermal innervation and thus
cannot be attributed to loss of innervation of tibial or-
igin. In fact, the near-total loss of innervation in the
tibial-innervated area after tibial, but not sural, axo-
tomy implies the existence of sharp borders between
the innervation territories of these two branches of the
sciatic. Although early sural-territory hyperalgesia could
reflect inflammation at the site of axotomy or more
proximally, its persistence for 5 months implicates
other mechanisms, such as the presence of axonal
sprouts with reduced thresholds for activation. Milder
hyperalgesia after SNI also develops in saphenous-
innervated skin.19 Because the saphenous and sciatic
do not share a common peripheral nerve but overlap
only at their shared border on the skin, and at termi-
nation within the central nervous system, central
mechanisms alone can produce hyperalgesia in skin ad-
jacent to areas innervated by damaged nerves.
We did not detect behavioral correlates of the con-
tralesional loss of PGP9.5⫹ neurites, although another
group studying SNI did,27 and contralateral hyperalge-
sia occurs in many unilateral nerve-injury models.6,28
Contralesional neurite loss in PHN patients also ap-
pears to be without contralateral behavioral correlate
because these patients almost never experience bilateral
pain after shingles.3 Surprisingly, though, the severity
of patients’ ipsilateral PHN pain correlates significantly
with the magnitude of contralesional neurite loss,3 rais-
ing the question of whether contralateral neurite loss
can influence the perception of ipsilateral pain.28
Our study also demonstrated ipsilateral, but not
contralateral, increases in PGP9.5⫹ neurites in adjacent
sural-innervated skin at 5 months after surgery (see Fig
4). This is presumably caused by collateral sprouting
from the adjacent uninjured sural nerve, because prox-
imal stumps of transected tibial nerves were tightly li-
gated to prevent regeneration. In contrast, contrale-
sional reductions in innervation were not followed by
sprouting. A prior study provides the physiological cor-
relate: it showed that although unilateral saphenous
nerve injuries produce bilateral reductions in distal
plasma extravasation, an ipsilateral-only return toward
normal levels of extravasation follows.10 However, oth-
ers have reported bilateral distal sprouting of motor ax-
ons after unilateral injury,11 so more study is needed to
determine if and when contralesional degenerative-like
changes trigger axonal regeneration.
Why might such profound contralateral effects of
nerve injury have gone so long undetected? Most in-
vestigators limit study to ipsilesional tissues only; if
contralesional tissues are studied, it is to provide “nor-
mative” data for control. One careful study of paw in-
nervation after partial sciatic nerve constriction in-
cluded samples from naive rats to provide a true
baseline against which to compare results from both
sides.29 The results showed a trend toward reduction in
contralateral PGP9.5⫹ innervation at 4 to 7 days after
surgery only, far less than we report. This may reflect
the relative innocuousness of their experimental injury
(loosely constricting, but not cutting, the nerve). In
contrast, nerve transection was used in the study that
reported profound bilateral loss of distal plasma extrav-
asation after one-sided injury.30 These results suggest
that physical separation of the injured nerve may be
necessary to trigger maximal contralateral effects.7
Transection per se, rather than prevention of ipsilateral
axonal regeneration by ligation, appears to be the trig-
ger for contralesional effects because these appear by 4
to 7 days after surgery,14,31 too soon for ipsilateral
axon regeneration to be a factor.
In our study, the restriction of contralesional
changes to tibial-innervated skin excludes a humoral
signal and suggests great spatial precision of the puta-
tive transmedian signal. Neurons of the tibial and sural
branches of the sciatic have somata within the same
sensory ganglia and spinal segments, and their central
axons terminate in the same regions, yet we found no
evidence of spread of transmedian changes from the
injured ipsilateral tibial branch to the contralesional
sural branch despite profound functional and anatom-
ical effects of tibial axotomy on the ipsilesional sural
branch.
At a minimum, our findings repudiate the common
practice of using tissues contralateral to a unilateral ma-
nipulation to provide control data. Failure to include
samples from naive or sham-operated animals as con-
trols can conceal contralesional effects, obscure or re-
duce the magnitude of ipsilateral changes, and poten-
tially lead researchers to erroneous conclusions.
A trend toward contralesional neurite loss was evi-
dent as early as 1 day after surgery, implying immedi-
ate activation of a putative transmedian signal that
travels at, or more rapidly than, fast axonal transport.
There is prior evidence for such signals. Unilateral for-
malin injection into rat hind paws induces robust bi-
lateral induction of phosphorylation of spinal cord
cAMP-responsive element-binding (CREB) within 1 to
2 minutes,9 and bilateral paw hyperalgesia can be in-
duced within 1 hour of unilateral nerve inflamma-
tion.27 Nonetheless, most distal contralesional effects
Oaklander and Brown: Bilateral WD 643
in our study and those of others are delayed by at least
several days.
In summary, we provide evidence of localized, long-
lasting loss of PGP9.5⫹ cutaneous innvervation con-
tralaterally to unilateral nerve injury, suggesting the ex-
istence of as-yet-unexplored rapid transcellular signals
linking homologous neurons on opposite sides of the
body. Corresponding changes also have been reported
in humans after various unilateral injuries, but in the
past these often went unrecognized by clinicians and
investigators alike. Our hope is to encourage use of ex-
perimental designs that facilitate detection of contrale-
sional effects and exploration of their clinical signifi-
cance.
This work was supported by the NIH (National Institute of Neu-
rological Disorders and Stroke, R01NS42866, A.L.O) and a Paul
Beeson Scholarship from the American Federation for Aging Re-
search (A.L.O).
We are indebted to I. Decosterd for performing surgeries and sensory
testing, to L. Zheng for morphometry, to Y. Chang and R. Gott for
assistance with data analysis, and to J. Campeti for editorial assistance.
References
1. Mitchell SW. Injuries of nerves and their consequences. New
York: Dover Publications, American Academy of Neurology
Reprint Series (1965 edition), 1872.
2. Kozin F, Genant HK, Bekerman C, McCarty DJ. The reflex
sympathetic dystrophy syndrome. II. Roentgenographic and
scintigraphic evidence of bilaterality and of periarticular accen-
tuation. Am J Med 1976;60:332–338.
3. Oaklander AL, Romans K, Horasek S, et al. Unilateral posther-
petic neuralgia is associated with bilateral sensory neuron dam-
age. Ann Neurol 1998;44:789 –795.
4. Watson CPN, Midha R, Devor M, et al. Trigeminal posther-
petic neuralgia postmortem: clinically unilateral, pathologically
bilateral. In: Devor M, Rowbotham MC, Weisenfeld-Hallin Z,
eds. Proceedings of the 9th World Congress on Pain. Seattle:
IASP Press, 2000:733–739.
5. Oaklander AL, Belzberg AJ. Unilateral nerve injury down-
regulates mRNA for Na⫹ channel SCN10A bilaterally in rat
dorsal root ganglia. Brain Res Mol Brain Res 1997;52:162–165.
6. Koltzenburg M, Wall PD, McMahon SB. Does the right side
know what the left is doing? Trends Neurosci 1999;22:122–127.
7. Chacur M, Milligan ED, Gazda LS, et al. A new model of
sciatic inflammatory neuritis (SIN): induction of unilateral and
bilateral mechanical allodynia following acute unilateral peri-
sciatic immune activation in rats. Pain 2001;94:231–244.
8. McLachlan EM, Janig W, Devor M, Michaelis M. Peripheral
nerve injury triggers noradrenergic sprouting within dorsal root
ganglia. Nature 1993;363:543–546.
9. Ji RR, Rupp F. Phosphorylation of transcription factor CREB
in rat spinal cord after formalin-induced hyperalgesia: relation-
ship to c-fos induction. J Neurosci 1997;17:1776 –1785.
10. Allnatt JP, Dickson KE, Lisney SJ. Saphenous nerve injury and
regeneration on one side of a rat suppresses the ability of the
contralateral nerve to evoke plasma extravasation. Neurosci Lett
1990;118:219 –222.
11. Rotshenker S, Tal M. The transneuronal induction of sprouting
and synapse formation in intact mouse muscles. J Physiol 1985;
360:387–396.
644 Annals of Neurology Vol 55 No 5 May 2004
12. Seltzer Z, Dubner R, Shir Y. A novel behavioral model of neu-
ropathic pain disorders produced in rats by partial sciatic nerve
injury. Pain 1990;43:205–218.
13. Attal N, Filliatreau G, Perrot S, et al. Behavioural pain-related
disorders and contribution of the saphenous nerve in crush and
chronic constriction injury of the rat sciatic nerve. Pain 1994;
59:301–312.
14. Sinnott CJ, Garfield JM, Strichartz GR. Differential efficacy of
intravenous lidocaine in alleviating ipsilateral versus contralat-
eral neuropathic pain in the rat. Pain 1999;80:521–531.
15. Noel F, Frost WN, Tian LM, et al. Recovery of tail-elicited
siphon-withdrawal reflex following unilateral axonal injury is as-
sociated with ipsi- and contralateral changes in gene expression
in Aplysia californica. J Neurosci 1995;15:6926 – 6938.
16. Navarro X, Verdu E, Wendelschafer-Crabb G, Kennedy WR.
Immunohistochemical study of skin reinnervation by regenera-
tive axons. J Comp Neurol 1997;380:164 –174.
17. Devor M, Schonfeld D, Seltzer Z, Wall PD. Two modes of
cutaneous reinnervation following peripheral nerve injury.
J Comp Neurol 1979;185:211–220.
18. Zimmermann M. Ethical guidelines for investigations of exper-
imental pain in conscious animals. Pain 1983;16:109 –110.
19. Decosterd I, Woolf CJ. Spared nerve injury: an animal model of
persistent peripheral neuropathic pain. Pain 2000;87:149 –158.
20. Tal M, Bennett GJ. Extra-territorial pain in rats with a peripheral
mononeuropathy: mechano-hyperalgesia and mechano-allodynia
in the territory of an uninjured nerve. Pain 1994;57:375–382.
21. Choi Y, Yoon YW, Na HS, et al. Behavioral signs of ongoing
pain and cold allodynia in a rat model of neuropathic pain.
Pain 1994;59:369 –376.
22. Thompson RJ, Doran JF, Jackson P, et al. PGP9.5: a new
marker for vertebrate neurons and neuroendocrine cells. Brain
Res 1983;278:224 –228.
23. McCarthy BG, Hsieh ST, Stocks A, et al. Cutaneous innerva-
tion in sensory neuropathies: evaluation by skin biopsy. Neu-
rology 1995;45:1848 –1855.
24. Hilliges M, Wang L, Johansson O. Ultrastructural evidence for
nerve fibers within all vital layers of the human epidermis. J In-
vest Dermatol 1995;104:134 –137.
25. Ma W, Bisby MA. Calcitonin gene-related peptide, substance P
and protein gene product 9.5 immunoreactive axonal fibers in
the rat footpad skin following partial sciatic nerve injuries.
J Neurocytol 2000;29:249 –262.
26. Bennett GJ, Xie YK. A peripheral mononeuropathy in rat that
produces disorders of pain sensation like those seen in man.
Pain 1988;33:87–107.
27. Erichsen HK, Blackburn-Munro G. Pharmacological characteri-
sation of the spared nerve injury model of neuropathic pain.
Pain 2002;98:151–161.
28. Gazda LS, Milligan ED, Hansen MK, et al. Sciatic inflamma-
tory neuritis (SIN): behavioral allodynia is paralleled by peri-
sciatic proinflammatory cytokine and superoxide production. J
Peripher Nerv Syst 2001;6:111–129.
29. Lindenlaub T, Sommer C. Epidermal innervation density after
partial sciatic nerve lesion and pain-related behavior in the rat.
Acta Neuropathol (Berl) 2002;104:137–143.
30. Kolston J, Lisney SJ, Mulholland MN, Passant CD. Transneu-
ronal effects triggered by saphenous nerve injury on one side of
a rat are restricted to neurones of the contralateral, homologous
nerve. Neurosci Lett 1991;130:187–189.
31. Araujo MC, Sinnott CJ, Strichartz GR. Multiple phases of relief
from experimental mechanical allodynia by systemic lidocaine: re-
sponses to early and late infusions. Pain 2003;103:21–29.