вход по аккаунту


Collateral reinnervation of sweat glands.

код для вставкиСкачать
Collateral Reinnervation of Sweat Glands
William R. Kennedy, MD, and Manabu Sakuta, MD
The collateral reinnervation of mouse sweat glands has been studied by a method that allows serial evaluation of the
course of reinnervation in intact animals. The method is based on the finding that the activation of secretion from
newly denervated sweat glands by pilocarpine or nerve stimulation is completely absent seven days after nerve section
but returns with reinnervation. These characteristics allowed serial detection of footpad sweat glands newly reinnervated by collateral sprouting of the remaining intact saphenous nerve after section of the sciatic nerve. The number of
saphenous innervated glands increased five- to sevenfold and the total saphenous sweat territory was greatly expanded
across the volar surface of the hind paw. There was less enlargement when the sciatic nerve was allowed to regenerate
and participate in reinnervation of sweat glands. None of the reinnervated glands were dually innervated by saphenous
and sciatic nerves, as is the case in normal glands. When the saphenous nerve was sectioned after saphenous collateral
sprouting was complete and after sciatic regeneration had innervated an apparent maximal number of glands, the
sciatic nerve reacted by advancing farther into the formerly enlarged saphenous territory and re-reinnervated many of
the glands. Collateral sprouting by sudomotor axons is more abundant and more widely dispersed than reported for
larger nerve fibers to skeletal muscle and to low-threshold mechanoreceptors. It more closely resembles sprouting of
nociceptive axons.
Kennedy WR, Sakuta M: Collateral reinnervation of sweat glands. Ann Neurol 15:73-78, 1984
Human [4, 13, 17) and mouse C14, 151 sweat glands
are apparent exceptions to the principles of denervation hypersensitivity proposed by Cannon 12). Following a complete nerve lesion, the denervated skin develops anhidrosis that is resistant to acetylcholine,
pilocarpine, methacholine, and epinephrine. This response is in contrast to that found in sweat glands of
the cat 118, 19, 211 and the low primate Lemur mongoz E201. These species are said to develop hypersensitivity to cholinergic agents within one week of denervation and to remain sensitive, although to a
lesser degree, for months.
The absolute insensitivity of the denervated mouse
sweat glands to cholinergic agents [15] provided a convenient system by which to study collateral sprouting
of sudomotor axons. Following single nerve lesions,
completely denervated sweat glands remain nonreactive to pilocarpine until reinnervated either by collateral sprouts from adjacent intact nerves or by the regenerated nerves. The progression of reinnervation can
be traced via periodic stimulation with pilocarpine to
determine the enlargement of the sweating territory of
the intact nerve and the increase in the density of
glands within the territory. This system to study
sudomotor function is of great interest, because, as
Sunderland [22) commented, “There have been no
comprehensive studies of regenerative processes in
sympathetic post-ganglionic nerve fibers.”
A method was devised to determine the exact number of
active sweat glands on the volar surface of the hind paws of
Swiss-Webster mice under various experimental conditions.
The alert mouse was placed in a plastic cylinder. The tail was
fixed to the outside surface to prevent withdrawal into the
cylinder and the hind limbs were held extended by skin clips
taped to the cylinder. Sweating was induced by a subcutaneous (s.c.) injection of pilocarpine nitrate, 5 mg per kilogram
of body weight, in 0.15 ml of water. Impression molds were
made of the plantar surface of the hind paws with a silicone
elastic material (Elasticon,Kerr Co, Romulus, MI) (0.2 ml of
base plus three drops hardener from a 22-gauge needle).The
hardened mold was removed, and the rounded indentations
produced by the emerging sweat droplets were counted
under a dissecting microscope. Baseline counts of sweat
glands activated by pilocarpine were made o n the hind paws
of each mouse upon entry into the study. The locarion of
each sweat droplet in the mold has been shown by serial
section of the same footpad to coincide with the position of
each sweat gland [l5].
From rhe University of Minnesota Hospitals, Box 187, Department
of Neurology, Minneapolis, MN 55455.
Received Feb 7, 1983, and in revised form May 23, 1983. Accepted
for publication May 30, 1983.
This report describes the changing responsiveness of
mouse sweat glands to pilocatpine during the first week
following denervation, the progressive reinnervation of
sweat glands on the mouse paw by collateral sprouting
of the saphenous nerve, and the effects on this collarera1 sprouting by the returning regenerated axons of
the previously sectioned sciatic nerve.
Address reprint requests to D r Kennedy.
The first objective was to determine the maximal number
of glands activated and the duration of the sweating produced
by pilocarpine stimulation. Impression molds were made after 4, 10,20,40, and 60 minutes, and then every 30 minutes
to 180 minutes. The number of glands counted were plotted
against time.
Next, the response of denervated sweat glands to pilocarpine was investigated in 12 mice. Under pentobarbital anesthesia (20 mg/kg), the sciatic nerve was sectioned about 9 mm
below the sciatic notch and a 2 rnm segment of nerve was
removed to delay regeneration. The saphenous nerve was
also sectioned high in the thigh. The distal end of each nerve
was then stimulated continuously with square pulses of 60
times the motor nerve threshold and 3 ms duration at ten per
second. Impression molds were taken to determine the number and distribution of glands activated. The mice were injected with pilocarpine s . ~ every
day for 10 days, and the
impression molds were taken to study the changes due to
A third group of mice was studied to learn the extent of
possible collateral sprouting of the saphenous nerve after sciatic nerve section, and the effect of sciatic nerve regeneration
on the process. In 10 mice the sciatic nerve was cut 9 mm
below the sciatic notch; in 10 others all branches of the sciatic
nerve were cut at the popliteal fossa. In the remaining 10 the
nerves were sectioned at the ankle level. The only glands
responsive to pilocarpine one week after the nerve section
were those later determined to be innervated by the saphenous nerve. All the mice were injected with pilocarpine S.C.
at weekly intervals, some for up to six months. The active
glands were counted and their positions plotted to follow the
course of collateral sprouting of the saphenous nerve into the
denervated areas. The molds were also observed for return of
sweating activity caused by regeneration of axons in the sciatic nerve.
To verify the results of the pilocarpine procedure, the degree of enlargement of the saphenous sweat territory was also
determined by nerve stimulation. At weekly intervals between six and twelve weeks, the nerve of some mice was
exposed and cut and impression molds were taken while the
distal nerve stump was stimulated. Sweat glands activated by
this procedure but not by pilocarpine 7 days later were innervated solely by the saphenous nerve. Activated glands that
remained responsive to pilocarpine for more than 7 days after
saphenous nerve section must have received axons from the
sciatic nerve. The progressive reinnervation by the regenerating sciatic fierve was charted weekly by making impression
molds after pilocarpine injection. At irregular intervals the
sciatic nerve of some mice was sectioned and then stimulated
to verify the extent of reinnervation of sweat glands, or of rereinnervation in the case of glands that had previously received collateral reinnervation by the saphenous nerve, which
had since been sectioned.
48 Hours
Minutes After Pilocarpine
Fig I . Decreased sensitivity to pilocarpine after denervation. The
percentage of jweat glands activated by pilocarpine declines progressively in the first 96 hours after total denerz'ation. The latency of onset increases and the duration of sweating decreases
during the first 48 hours and then remains cow.rtant in theJw
glands that can stiN be activated.
or one footpad. As shown previously ClS], all the saphenous-activated glands were also activated by the sciatic nerve. Subcutaneous pilocarpine injection activated over 909% of the glands activated by nerve
stimulation. The maximum number of glands was active 10 to 20 minutes after pilocarpine was injected.
Eighty percent continued to sweat longer than two
hours (Fig 1).
Denervation Experiments
After the sciatic and saphenous nerves were cut, the
number of sweat glands activated by pilocarpine declined daily and reached zero by day 7. This finding
indicates that every sweat gland in the hind paw is
innervated by either the sciatic or the saphenous nerve.
During this period after denervation, the latency of
sweating after pilocarpine injection progressively increased (see Fig 1). Precise examination of the molds
indicated that there was a tendency for the denervated
sweat glands that were activated at a shorter latency (20
to 40 minutes after pilocarpine injection) to be active
for the longest duration (100 to 120 minutes after
pilocarpine injection). The threshold of the denervated
glands was increased, as indicated by the decreasing
number of glands responsive to decreasing concentrations of pilocarpine. Hypersensitivity was not observed
during a six-month period so long as nerve regeneration was prevented.
Normal Glands
Reinnervation of Sweat Glands
In the intact mouse, electrical stimulation of the sciatic
nerve activated all sweat glands in the hind paw. Saphenous nerve stimulation usually activated only a few
sweat glands on digits 1 and 2, rarely a gland on digit 3
In group 3 one week after the sciatic nerve was cut,
only those sweat glands innervated by the saphenous
nerve were activated by pilocarpine injection (Fig 2A).
Two weeks after the sciatic nerve lesion, the saphenous
74 Annals of Neurology
Vol 15 No 1 January 1984
Fig 2. Findings after sectioning of the sciatic nerve in the thigh.
(A)Subcutaneous injection of pilocarpine one week after section
activated only the innervated sweat glands, those supplied by the
saphenous nerve, as indicated by the dots. (B) Two weeks afier
sectioning, saphenous innervation had already extended t o additional glands in a wider distribution. (C) By six weeks after sectioning, the saphenous tewitoy had expanded over much of the
paw to five to w e n times the number of glands originally innervated. saphenous innervation of the pilocarpine-activated glands
was proved when the same glands were activated by electrical
stimulation and then became refvactory seven days after saphenous nerve section. (Note: The saphenous nerve remained intact.
The dots and crosses mark the location of individual activated
sweat glands as detected by the impression of their secreted sweat
droplets in the Silastic mold.)
sweat territory began to increase in the density of active sweat glands innervated and in total area (Fig 2B).
By six weeks (Fig 2C) the number of saphenous reinnervated glands was maximal. Thereafter the active
sweat glands remained about the same in number and
location until about twelve weeks, when some contraction of the territory occurred. Note that the two glands
Fig 3. Findings after distalsectioning of the sciatic nerve and its
branches. Six wee& after sectioning, sweat glands were excited by
electrical stimulation of either saphenous (dots) or sciatic
(crosses) nerve. Enlargement of the saphenous territory was less
than that shown in Figuve 2C because ofthe competition from
regenerating sciatic axons. The variability seen here in A and B
presumably depends on the number of sciatic nerve axons that
reach the paw. (See Note accompanying Figure 2.)
on digit 5 in Figure 2B are not seen in Figure 2C. The
amount of contraction varied in different animals but
was estimated never to exceed 10%. The contraction
was usually observed on the fourth or fifth digits, which
were on the outskirts of the expanded territory and had
only a few saphenous-innervated glands. All the glands
responding to pilocarpine also reacted to saphenous
nerve stimulation. Furthermore, after the saphenous
nerve was cut at this stage, every sweat gland became
inactivated within 7 days. From this evidence the newly
activated sweat glands were judged to be collaterally
innervated only by the saphenous nerve. The total saphenous area had tripled or quadrupled, and the number of innervated sweat glands had increased five- to
When the sciatic nerve was allowed to regrow, the
degree of sprouting of the saphenous nerve was inhibited. This was evidenced by the decreased number
of glands activated by saphenous stimulation and the
relatively small area that became anhidrotic after the
saphenous nerve was cut (Fig 3A, B). The inhibition
was greatest in the mice with the most distal original
sciatic lesion. These axons regenerated into the paw in
the shortest time, between two and three weeks, before sprouting of saphenous axons had reached its
Kennedy and Sakuta: Reinnervation of Sweat Glands
enlarged saphenous territory (Fig 4 B ) so that several of
the glands had been re-reinnervated. The first reinncrvation had been by the saphenous nerve, the second by
the sciatic nerve. Full reinnervation of all glands was
never obtained.
Fig 4.(A) Typical reinnewation pattern twelve weeks after distal sciatic nerve section shous the location of glands innervated by
the saphenous newe (dots) and the regenerated sciatic nerve
(crosses). (Bi The same paw twenty weeks after sciatic sectioning
and eight weeks after .taphenous sectioning. Some of the glands
originally innervated by the sciatic nerve, then reinnerrated by
the saphenous netwe as in ( A ) ,have now been re-reinnervated by
the sciatic nerve. (See Note accompanying Figure 2.1
fullest extent. Those glands that remained active after
saphenous denervation were activated by sciatic nerve
stimulation but became inactive one week after the
sciatic nerve was cut again. These results provided
proof of sciatic reinnervation.
Weekly testing with pilocarpine showed that collateral reinnervation by the saphenous nerve and reinnervation by the regenerating sciatic nerve plateaued after
six to seven weeks and remained constant. We did not
attempt to measure the number of glands that the saphenous territory lost to retraction and the sciatic territory reclaimed by re-reinnervation. At twelve weeks
the saphenous and sciatic nerves were stimulated one at
a time in a few mice. There was almost no overlap
between the two innervation areas. These findings contrasted with those in normal mice, in which all saphenous-activated glands were also activated by the sciatic
In several mice the saphenous territory was evaluated at twelve weeks by stimulation of the cut nerve.
The glands in the regenerating sciatic nerve were determined by pilocarpine activation one week later (Fig
4A). These mice were reevaluated at twenty weeks by
means of sciatic nerve stimulation. It was found that the
sciatic reinnervation had extended into the previously
76 Annals of Neurology
Vol 15 No 1 January 1984
The mouse proved to be a good animal model to study
regeneration of sudomotor axons because denervated
mouse sweat glands, like those of humans, cease tso
respond to pilocarpine, acetylcholine, and epinephrine.
The only glands activated by pilocarpine after the sciatic nerve was cut were those innervated by intact axons and their collateral sprouts. This apparent exception to the law of denervation hypersensitivity 121
made it possible to follow the progress of collateral
reinnervation and of delayed regeneration in the samc
During the weeks following sciatic nerve section, an
increasing number of sweat glands responded ti.)
pilocarpine. In the absence of other intact nerves to the
paw, the source of reinnervation must have been collateral sprouting from the intact saphenous nerve. This
sprouting resulted in greater density of glands withiri
the saphenous territory and expansion of the saphenous sweat area far beyond its original boundary onto
other sweat pads and digits. The source of innervation
was verified by activation of all pilocarpine-sensitive
glands by electrical stimulation of the saphenous nerve
and by the disappearance of activity one week after the
nerve was severed. In normal mice the saphenous
nerve innervates from nine to twenty sweat glands. Six
weeks after sciatic nerve section, the number of active
sweat glands had increased to a maximum, between
sixty and one hundred, for a ratio of 5 : 1 to 7 : 1, slightly
greater than the fourfold increase in motor unit size
reported to follow partial denervation of skeletal muscle [23]. It was not possible to determine the increased
ratio per sudomotor axon, because we have not yet
done morphological studies to determine whether the
saphenous nerve normally innervates its territory by
only a few axons that each branch to several glands, by
one axon per gland, or by several axons per gland. The
retraction of the enlarged saphenous territory at about
twelve weeks never remotely approached retraction to
the original territory, which has been reported in
sprouting motor axons to skeletal muscle [ I , 61.
In the mouse, additional reinnervation must have
occurred that was not reflected in the sweat gland
counts. Prior to sciatic nerve section, all glands in the
saphenous territory could also be activated by stimulation of the tibial nerve [15}. Thus, they were coinnervated by the two nerves. After nerve section the
glands with the surviving saphenous supply could still
be activated by nerve stimulation or pilocarpine, but
the droplets produced were slightly smaller. This
finding suggests that these partially denervated glands
had curtailed their production of sweat. The first step
in reinnervation presumably was sprouting from the
surviving axons to vacated receptor areas within the
same glands. The time required for this initial regrowth
was not measurable but was probably not more than
two weeks, because by that time droplet size had increased and adjacent glands within the confines of the
saphenous nerve territory but not originally innervated
by that nerve had begun to respond to pilocarpine. The
reinnervation eventually extended outside the saphenous territory to digits and pads deep within the tibial
nerve autonomous zone. W e have dissected the normal
saphenous nerve and found that it proceeds distally
along the medial dorsal aspect of the paw, giving off
long branches that reach the dorsum of the digits.
These proceed to the volar surface by encircling the
proximal portion of the most medial two or three digits. Therefore, nerve sprout growth from the normal
saphenous territory must progress distally to the sweat
pad on the tip of the digit but also proximally to the
sweat glands on the pads of the paw, probably within
existing endoneurial channels. Where the sprouts grew
into digits not normally innervated in part by the saphenous nerve, the pathway must have been direct and
outside of existing endoneurium. The exact path taken
by the collateral sprouts to the reinnervated sweat
glands must be evaluated by morphological studies.
The described enlargement of the sweating territory
of the intact saphenous nerve was restricted when sciatic nerve axons successfully regenerated into the paw.
This usually occurred in the mice with more distal lesions. Even in these mice the moderate expansion of
the saphenous territory prevented the regenerating sciatic axons from reaching their full reinnervation potential. This restriction was evidenced when severance of
the saphenous nerve at twelve weeks was followed by
further expansion of sciatic axons into the now denervated saphenous territory. The paucity of coinnervation of individual sweat glands by the two
nerves, as found in normal mice (151, is probably
caused by the full occupation of all receptor sites by the
saphenous nerve before the sciatic axons grew into the
paw. Reinnervated mammalian muscle fibers, in contrast, are often doubly innervated by mature collateral
sprouts and by regenerating axons (121, at least for a
short time (11.
The methods we employed to study sudomotor axons have some advantages over methods used to study
other types of degenerating nerves. The state of
sudomotor reinnervation could be evaluated at desired
intervals without killing the animals or disturbing the
reinnervation process. Determination of successful
reinnervation by functional testing with pilocarpine
avoided the possibility of mistaking regenerating nonmyelinated axons conveying pain, sensation, or
vasomotor impulses for sudomotor axons, as might occur in morphological studies. Furthermore, reinnervation could be verified by electrical stimulation. If morphological studies are desired, small groups of
reinnervated glands can be removed at biopsy and
studied at selected times after functional reinnervation
has occurred. We have not yet determined whether
electrical stimulation with the currents described
causes damage to axons. This possibility was not a factor in these experiments, because the nerves were
severed before stimulation of the distal stump to prevent any possible reflex activation of the adjacent
The degree of sweat area enlargement by collateral reinnervation that we observed suggests that
sudomotor axons are not governed by the constraints
that restrict sprouting of other types of nerve fibers.
Although reinnervation by nerve fibers conveying different types of information has been evaluated by
methods that are adapted to the characteristics of the
particular system and therefore different from ours,
some comparisons are possible. Intact nerve fibers
to low-threshold mechanoreceptors display minimal
sprouting into adjacent denervated skin in salamanders
and adult rats {5]. Sprouting is more prolific in immature rats but remiuns within the same dermatome (91.
Regenerating axons, however, do not remain within
dermatomal boundaries [lo]. In muscle residual alpha
motor nerve fibers to partially denervated rat and human skeletal muscle collaterally reinnervate muscle
fibers by short terminal or nodal sprouts C3J, thereby
increasing the density of the surviving motor units
[16]. The total area of the motor unit, as precisely
mapped by histochemical depletion methods, does not
appear to be significantly increased [ll]. There is evidence, however, that in some circumstances alpha
motor axons have the capacity to send longer collateral
sprouts outside the parent motor unit territory C231.
Nerve fibers conveying nociceptive stimuli most
closely resemble sudomotor axons, because sprouts
from these axons also grow into the denervated autonomous sensory zone of neighboring severed peripheral
nerves C7, 241 or dermatomes (81.
W e suggest that the sprouting features of each functional or morphological category of nerve fiber will be
found to be characteristic of those particular fibers
within a species, with additional, perhaps unsuspected,
features becoming evident as the experimental situation is altered. The ability of the unmyelinated sudomotor axons to reinnervate sweat glands at a distance
from their normal sweat territory probably reflects the
fact that sweating must be diffuse to accomplish its
purpose of cooling. In this sense there are similarities
between the slow-conducting unmyelinated sudomotor
system and the warning aspects of the small-axon
nociceptive system.
Kennedy and Sakuta: Reinnervation of Sweat Glands
Presumably the rigid constraints imposed on collateral sprouting of the larger nerve fibers to lowthreshold mechanoreceptors and to muscle exist to
preserve the precise spatial organization of peripheral
organ innervation necessary to retain correct recognition of objects by touch and exact coordination of
The responsiveness of denervated sweat glands to
chemical transmitters appears to depend on the species
tested. The lower primate Lemur mongoz [20) and cats
118, 19, 21) are reported to be hypersensitive to cholinergic drugs and to retain their responsiveness for
several weeks after denervation. The reason for the
occurrence of hypersensitivity in these species but not
in humans or mice is unknown. The variations in the
experimental methods used by the different authors
preclude useful comparison of sweat gland characteristics.
Supported in part by grants from the Minnesota Medical Foundation
and the American Diabetes Association, Minnesota Chapter, and by
Grant AM20516-01 from the National Institutes of Health
The authors express their gratitude to D r Kwon S Yoon, &chard
Landis, Kevin Alto. and Lucy Davenport for assistance in the experimental work, and to Carole Halper for preparation of the manuscript
1. Brown MC, Holland RL, Hopkins WG: Motor nerve sprouting.
Ann Rev Neurosci 4:17-42, 1981
2. Cannon WB: A law of denervation. Am J Med Sci 198:737750, 1939
3. Coers C, Woolf A L The Innervation of Muscle. Oxford, Blackwell, 1959
4. Coon JM, Rorhman S: The sweat response to drugs with nicotine-like action. J Pharmacol Exp Ther 73:l-11, 1941
5. Cooper E, Diamond J: A quantitative study of the mechanosensory innervation of the salamander skin. J Physiol (Lond)
264:695-723, 1977
6. Dennis MJ, Yip JW: Formation and elimination of foreign synapses on adult salamander muscle. J Physiol (Lond) 274:293310, 1978
7. Devor M, Schonfeld D, Seltzer 2, Wall PD: Two modes of
78 Annals of Neurology Vol 15 No 1 January 1984
cutaneous reinnervation following peripheral nerve injury. J
Comp Neurol 185:2 11-220, 1979
8 Diamond J: The recovery of sensory function in skin after peripheral nerve lesions. In Gorio A, Millesi H , Mingrino S (eds):
Posttraumatic Peripheral Nerve Regeneration: Experimental
Basis and Clinical Implications. New York, Raven, 1981, pp
9. Diamond J, Jackson PC: Do cutaneous nerves sprout in the
mammal? J Physio1280:52P-53P, 1978
10. Diamond J, Jackson PC: Regeneration and collateral sprouting
of peripheral nerves. In Jewett DL, McCarroll H R Jr (eds):
Nerve Repair and Regeneration: Its Clinical and Experimental
Basis. St Louis, Mosby, 1980, pp 115-129
11 Edstrijm L, Kugelberg E: Histochemical composition, disrribution of fibers and fatiguability of single motor units. J Neurol
Neurosurg Psychiatry 31:424-433, 1968
12 Guth L Neuroquscular function after regeneration of interrupted nerve fibers into partially denervated muscle. Exp Neurol
6:129-141, 1962
13. Kahn D, Rothman S: Sweat responses to acetylcholine. J Invest
Dermarol 5:43 1-444, 1942
14. Kennedy WR, Sakuta M: Denervation and reinnervation of
sweat gland (abstract). Soc Neurosci 8:27, 1982
15. Kennedy WR, Sakuta M: Multiple innervation of sweat gland.
Neuroscience (in press)
16. Kugelberg E, Edstrom L, Abbruzzese M: Mapping of motor
units in experimentally reinnervated rat muscle. J Neurol
Neurosurg Psychiatry 33:3 19-329, 1970
17. List CF, Peer MM: Sweat secretion in man. 111. Clinical observations on sweating produced by pilocarpine and Mecholyl. Arch
Neurol Psychiatry 40:269-290, 1938
18. Nakamura Y, Hatanaka K: Effect of denervation of the car’s
sweat glands on their responsiveness to adrenaline, nicotine and
Mecholyl. Tohuku J Exp Med 68:225-237, 1958
19. Reas HW, Trendelenburg U: Changes in the sensitivity of the
sweat glands of the cat after denervation. J Pharmacol Exp Ther
1561126-136, 1967
20. Sakurai M, Montagna W: Observation of the eccrine sweat
glands of Lemur mongoz after denervation. J Invest Dermatol
44187-92, 1965
21. Simeone FA, Mentha C, Rodrigues HA: Responsiveness of
sweat glands after denervation by preganglionic ramisection,
ganglioqectomy and peripheral nerve section. Am J Physiol
165:356-364, 1951
22. Sunderland S: Nerves and Nerve Injuries. Second edition. Edinburgh, London, and New York, Churchill Livingstone, 1978
23. Thompson W, Jansen JKS: The extent of sprouting of remaining
motor units in partly denervared immature and adult rat soleus
muscle. Neuroscience 2:52 3-53 5 , 1977
24. Weddell G, Guttmann L, Gurmann E: The loca! extension of
nerve fibres into denervated areas of skin J Neurol Psychiatry
4:206-225, 1941
Без категории
Размер файла
625 Кб
reinnervation, gland, collateral, sweat
Пожаловаться на содержимое документа