вход по аккаунту


The nerve supply of the vertebral column and its associated structures in the monkey.

код для вставкиСкачать
Department of Anatomy, Stanford University,
Stanford, California
Information in the literature concerning the intrinsic nerve
supply of the vertebral column and its associated structures
deals primarily with the effort to provide details essential to
clinical studies, and gives no data for possible investigations
of a functional nature related to its role in posture and locomotion. Research seems to have been limited to man, and
no single study is complete. Furthermore, the individual
reports vary. The origins of the nerves supplying the various
individual structures concerned with the spine and its function have never been stated, with the exception of nerves
to the posterior longitudinal ligament and dura mater, the
meningeal ramus (B.N.A.), “recurrent” or “sinu-vertebral”
nerve. Other problems remain to be solved, such as segmental
overlap, the variety and frequency of terminal receptors, and
the size and myelination of the nerve fibers.
I n order to understand both the functional and clinical
significances of this system of nerves, it seems essential to
have more complete information concerning their origin,
distribution, and termination.
Monkeys (rhesus and Cynomolgus) were chosen, because
they are being used in this laboratory in other studies related
JWNE 1956
125, NO. 2
to the spine, are commonly used in neurophysiological studies,
and because they can be employed for the demonstration of
nerves by the intravital methylene blue technique.
A combination of myelin and intravital methylene blue
stains was used to study the nerves in 11rhesus and Cynomolgus (Java) monkeys. I n 4 monkeys the whole vertebral column
was removed with rib stumps and spinal muscles attached and
was fixed in a fluid containing 10% formalin, 1%calcium
chloride, and 2.5% cobaltous nitrate. Following decalcification in a mixture of equal parts of 50% sodium citrate and
50% formic acid, three or more successive segments from the
central regions of the cervical, thoracic, and lumbar spine
were removed. They were then infiltrated with gelatin, refixed in formalin-calcium-cobalt fluid and cut serially in
transverse or sagittal planes at 50-75 p on the freezing
microtome. These serial sections were mounted on slides
with 1%gelatin solution by chilling, then fixed in cold formalin-calcium-cobalt fluid. Staining was carried out with
Sudan black B in accordance with the method of McManus
( '46). This technique stains myelin a dark gray-green. The
sections were employed to trace larger branches of the nerves
to their destinations in the tissues to be examined. Other
investigators have used the Masson stain to demonstrate
larger nerves (cf. Gardner, '48); however, the Sudan stain
was found to be much superior to a triple stain in this
material. Very thick sections are quite transparent and it
is easy to trace large nerves, all of which have at least a
few myelinated axons. Even individual myelinated fibers
can be followed in muscle, bone and other connective tissues.
Seven other monkeys were perfused intravenously with
methylene blue under nembutal anesthesia, following a method
of Feindel, Sinclair and Weddell ( '47). A minimum of three
hours was necessary to produce adequate staining of nerves
and their terminations. After perfusion the spines were
quickly removed and the following structures were dissected
free : anterior and posterior longitudinal and interspinous
ligaments, articular capsules, annulus fibrosus, periosteum
and ligamenta flava. Fixation for 6 or more hours in cold
8% ammonium molybdate was followed by a brief wash in cool
isotonic sodium chloride. Full thickness (in most cases) or
freehand cuts of the tissues were flattened between slides,
rapidly dehydrated in absolute alcohol, and cleared in xylol,
followed by methyl salicylate.
This method has the advantages of making the most delicate
nerve endings visible and of demonstrating the entire intramupal plexus in the tissues so treated. Such a technique is
certainly preferable to reconstruction of serial sections.
Fig. 1 Profile of three articulated segments of the lumbar region, showing
branches of two successive dorsal rami which supply articular capsules. The lower
joint receives nerves from the nearby dorsal ramus as well as from a r m u s one
segment cranial.
T h e distribution. of dorsal and ventral rami of t h e spinal
nerves. I n the lumbar and thoracic parts of the spine, the
spinal ganglia lie partly inside but largely outside the intervertebral foramina. The dorsal ramus arises from its spinal
nerve just distal to the ganglion or from its dorsal surface
and divides almost immediately into several branches which
are directed mainly dorsally, but also cranially and caudally.
One most medial branch arising from the dorsal ramus
passes under the corresponding transverse process, then
Fig. 2 Dorsal and ventral ramus branc,hes in the thoraeie region. A cranially
directed branch of the dorsal ramus supplies deep oblique muscles and an articular
capsule cranial to its level of origin. Another dorsal ramus branch supplies a
joint one segment caudal, continuing into dorsal muscles. Ventral rami (not
labelled) have been severed. Autonomic rami, two or more in number, give off
branches to periosteum and areolar connective tissue on the surface of the intervertebral disk. The anterior longitudinal ligament and nearby periosteum receive
nerves which arise from the sympathetic ganglia.
cranially and medially to supply the deep oblique muscles
(rotators, multifidus, semispinalis). Two or more different
branches of the dorsal ramus are directed dorsally and then
caudally to supply the lateral parts of the sacrospinalis and
some of the deep oblique muscles. The medial ascending
branches of one spinal nerve invariably overlap the more
lateral descending branches from the cranially adjacent spinal
nerve. Hence, deep oblique muscles are supplied by cranially
directed nerves from the same segmental level and by others
from a level immediately cranial, while parts of the sacrospinalis derive their nerves from the segment with which they
are associated (figs. 1 and 2).
At cervical levels the dorsal rami are similar in that an
overlap of adjacent segments occurs and that cranially and
caudally directed branches are present, but division of the
dorsal ramus occurs farther dorsally, usually lateral to the
articular process (figs. 3 and 18). In addition, longitudinal
communications are found between adjacent dorsal rami of
cervical nerves along a line connecting their origins from
spinal nerves (fig. 3).
Branches of ventral rami are single at thoracic, usually
multiple at cervical, and variable at lumbar levels, depending
apparently upon the nature of their contributions to the
lumbosacral and cervical plexuses. All are directed somewhat caudally to their destinations in the plexuses (figs. 1-3).
The autonomic rami, sympathetic chain, and paravertebral
autofiomic plezuses. Gray and white rami communicantes
arising from the spinal nerves or from the spinal ganglia of
the monkey vary in number at the different levels. Generally
there is but one lumbar (gray?) ramus communicans for each
segment, but usually two or three arise from each thoracic
and cervical segment. The manner in which these rami contribute to an autonomic plexus connected to ventral and dorsal
spinal nerve rami and to the sympathetic and dorsal root
ganglia varies according to the representative spinal level
At the lumbar level usually only one ramus connects a spinal
to a sympathetic ganglion, Although fibers within this nerve
vary in size and degree of myelination, it seems probable
that the single communication is a gray ramus, containing
some myelinated somatic axons. Two or more branches of this
autonomic ramus invariably are given off its medial surface
to the periosteum of the vertebral body and to the annulus
fibrosus, as the nerve circles the body. I n addition, branches
Fig. 3 Branches of dorsal and ventral rami in profile a t the mid-cervical region.
Dorsal rami communicate by vertically-running nerves. Division of the dorsal
rami occurs as they lie opposite the cervical articulations; a t this point a t least
two small branches penetrate the joint capsule. The most, caudal branch supplies
'deep oblique muscles and gives off a filament t o the joint a segment caudal to its
origin. I n addition a small branch may arise from the dorsal ramus near its
origin and supply an articulation nearby or caudal.
The vertebral plexus is shown arranged around the vertebral artery, and is connected to each spinal ganglion by two or more rami communicantes. Branches of
this plexus supply periosteum, the surface of the annulus fibrosus, and the anterior
longitudinal ligament.
are given off near its attachment to the spinal ganglion to
join each branch of the ventral and dorsal spinal rami and to
large branches of segmental arteries, while others recur
into the intervertebral foramen ventral and dorsal to the
spinal ganglion to supply the periosteum, dura mater, posterior longitudinal ligament, and to enter deficiencies in the
vertebral body. The result, though stated here rather simply,
is a single autonomic ramus with a plexiform multitude of
small nerves communicating with ventral and dorsal spinal
rami and supplying nearby associated vertebral structures
with visceral and somatic nerves.
The plexus at thoracic levels is basically similar to that of
the lumbar segments just described, but the paravertebral
autonomic plexus is even more complex, since two or three
rami connect a spinal with a sympathetic ganglia, all of the
rami intercommunicating, receiving branches from or giving
off small branches to the periosteum, bone, costotransverse
joint, posterior longitudinal ligament, segmental vessels and
dura mater (fig. 4).
At mid-cervical levels this autonomic plexus is similar to
those at more caudal levels, with the exception of the source
of some of its fibers and its communications. The cervical
spinal ganglia lie on the cranial surface of the transverse
processes, covering the dorsal half of the foramina intertransversarii. Ventral to the ganglia and running longitudinally
through the foramina are the vertebral artery and veins,
surrounded by a plexus of autonomic nerves which has as its
source inferior portions of the cervical sympathetic chain.
According to Riegele (monkey and ape, '26), Siwe (man, '31),
and Gardner (man, '43), this plexus of autonomic fibers associated with the vertebral vessels arises from the inferior
cervical or stellate ganglia (the vertebral nerve) and from
the subclavian plexus (the vertebral plexus). Siwe states
that the vertebral plexus receives gross communications from
the cervical spinal roots and that additional gross communica-
tions occur with the vagus and glossopharyngeal nerves ; the
latter also are said to have other connections with ganglia of
the cervical sympathetic chain,
Fig. 4 Horizontal seetion at a thoracic level showing the main branches of
the spinal ganglion and of the dorsal ramus. A paravertebral nerve plexus is
shown (see text), made up of communications between the spinal and sympathetic
ganglia, with branches to the longitudinal ligaments, dorsal and ventral (intercostal) rami, periosteum, spongy bone of the rib, the vertebral body and arch,
and t o costotransverse and intervetebral joints. The sources of the meningeal
ramus (<‘recurrent” nerve) are shown. Muscular branches of the dorsal ramus
give off nerves to the intervertebral joints; these correspond to articular nerves
in figure 2.
A survey of the vertebral autonomic plexus was not made,
but certain pertinent additions to the above can be supplied for
the monkey. As the plexus courses cranially in the foramina
it receives not one but several communicating branches from
the adjacent ventral surfaces of the spinal ganglia (figs. 5
and 9). Branches of the plexus thus enhanced are peripherally
distributed : (a) cranially and caudally within the plexus
(fig. 10) ; (b) to the anterior longitudinal ligament and the
periosteum of the ventral aspect of the body of the vertebra
via a branch directed through the longus colli muscle, communicating with the sympathetic chain and runing medially
to the body; (c) to the lateral portion of the vertebral body
and its periosteum through fibers directed medially from the
plexus; ( d ) to the dura mater, periosteum and the posterior
longitudinal ligament by branches recurring into the intervertebral foramen; (e) to the dorsal spinal rami by several
branches; and, ( f ) to the ventral rami of the spinal nerve
by one or more branches directed lateralward. Through these
connections of the plexus with spinal and sympathetic rami
routes are provided for connections of the intrinsic structures
of the vertebral column with somatic and visceral nerves
(figs. 3 and 5 ) .
T h e anterior longitudinal ligameat. I n the lumbar and
thoracic regions this ligament is supplied by direct branches
of the sympathetic ganglia which enter its lateral aspect both
cranial and caudal to the intervertebral disk. The cervical
portion of the ligament is supplied by nerves arising from
the vertebral plexus, coursing ventrally through the longus
colli muscle, and then medially on the surface of this muscle
to the ligament (figs. 2-5 and fig. 11).
I n cleared methylene blue preparations the ligament can
be seen to have a plexus of nerve fibers throughout its substance, though situated predominantly near its deep surface.
Nerves entering its lateral border adjacent to an intervertebral disk immediately break up into fibers which pass
cranially, caudally and medially. Those directed medially
overlap those of the opposite side by at least half the width
of the ligament, while those running longitudinally pass and
overlap others entering from cranial and caudal. All regions
in the anterior longitudinal ligament are thus supplied by
axons derived from the same level, and from segments immediately adjacent. There is no doubt of such an arrangement; in specimens prepared by this technique it is possible
Fig. 5 Innervation of related cervical vertebral structures in transverse section.
Muscular branches of the dorsal ramus supplying the articular capsule correspond
t o those in figure 3. The vertebral plexus is seen within the transverse foramen
(unlabelled). Communications of the plexus are seen with the spinal ganglion,
dorsal and ventral rami, and the sympathetic trunk (and via this branch to the
periosteum and marrow of the vertebral body and the anterior longitudinal ligament). Other branches are directed medially to the periosteum and spongy bone
of the body and via the meningeal ramus t o the dura mater and posterior longitudinal ligament.
to follow single nerve fibers from their point of entrance to
their terminations, and by this means to determine the origin
of neurons intrinsic within the ligament.
No critical study was made of nerve fiber size. Only fibers
of less than 10 F were seen. Those of less than 5 p were by f a r
the more predominant. These axons bifurcate repeatedly, a
Fig. 6 Two fibers less than 5 fi in diameter in the posterior longitudinal ligament near its lateral edge. These axons dichotomize repeatedly and each gives
rise to a number of arborizing endings with long, tapering, fllamentous tips. This
type of termination was found i n the capsule of vertebral joints, lumbodorsal
fascia, and the longitudinal ligaments. Drawn from a whole mount of the ligament, methylene blue stain. X 200.
single neuron having perhaps 5-15 endings with delicate,
arborizing branches, their tips tapering into invisibility
(fig. 6). Such endings are common in other connective tissues
of the body, and probably are pain receptors. They have been
described by Sfameni ('02) in adipose tissue and in periodontal membrane by Lewinsky and Stewart ( '36). Other
small, unmyelinated fibers, similarly bifurcated, more often
beaded, end in tiny terminal enlargements, singly or in groups.
These probably also are a type of pain receptor. Fine, beaded,
unmyelinated fibers continue along vascular branchings to
form the usual network of vasomotor axons to arterioles.
Some of these continue into bone.
Another, slightly larger, type of nerve fiber is slightly
myelinated. These tend t o follow vessels and enter deficiencies
in the cortical bone of vertebral bodies, almost immediately
losing their myelin sheath. The last two types of nerves
resemble those described by Huntz and Richins ('45) which
in the dog and cat enter the marrow of spongy bone.
T h e posterior longitudinal ligament. This ligament is described in other reports as receiving its nerve supply from a
branch of the spinal nerve which recurs into the ventral part
of the intervertebral foramen, the meningeal ramus, ' ' recurrent" or "sinu-vertebral" nerve. Previous descriptions in
man, by Jung and Brunschwig ( '32), Roofe ( '40), Wiberg
( '49) indicate either that the ligament is supplied by a branch
of the spinal nerve, by a sympathetic ramus, or by both
(Hovelacque, '25). There appears to be considerable variation in the source of nerves to this ligament; however, the
material examined in this study indicates that it has a dual
nerve supply, and that fibers of both somatic and visceral
sources are constantly to be found in it. It is the rule to find
not one but two recurrent nerves entering at each foramen,
one from the paravertebral autonomic plexus described earlier, and another arising from a spinal ganglion, a spinal
nerve, or from either ventral o r dorsal spinal roots, or even
from a combination of two of these sources (figs. 12,4 and 5).
As these nerves reach the lateral portion of the posterior
longitudinal ligament branches are, as in its anterior counterpart, directed cranialward, caudalward, and medialward, and
a similar overlap exists between adajacent vertebral nerve
segments. The distribution of axons within the ligament, the
fiber size, and the types of their termination are similar to
those in the anterior longitudinal ligament. Again, the myelinated fibers of moderate diameter seem destined for bone,
and nearly all of them enter a large central osseous deficiency
on the dorsal surface of the vertebral body, after being directed cranialward or caudalward a half segment from their
entrance at the foramina.
Fig. 7 A horizontal section through the lateral part of the posterior longitudinal
ligament (P.L.L.), at a point where it fuses with t,he annulus fibrosus (A.F.).
For the most part vessels and nerves are found only in the ligament, although in
reaching their terminations they may lie within the most superficial lamellae of
the annulus fibrosus. Free axon terminations are limited to the posterior longitudinal ligament. Beaded vasomotor fibers can be seen supplying the arterioles.
Drawn from a cleared free-hand cut of the ligament and intervertebral disk.
Methylene blue stain. x 500.
The axons tend to follow blood vessels, myelinated fibers
and a few smaller unmyelinated ones accompanying vessels
to bone; small, beaded autonomic axons are found in simple
plexuses in and near vessels, and other small unmyelinated
fibers, apparently restricted to a sensory capacity, dichoto-
mize repeatedly and are directed to the surface of the ligament facing the vertebral canal. At or very near this surface
these fibers end with minute terminal enlargements, singly
or in groups (figs. 6, 7 and 13-15).
T h e dura mater. Nerves destined for the dorsal part of
the dura mater arise exclusively as branches of the autonomic
plexus arranged around the origins of spinal nerve rami. A
large branch from this plexus usually enters the intervertebral
foramen dorsal to the spinal ganglion, there encountering the
external surface of the dura mater. Small branches of these
nerves also are given off to the surface of the ligamenta flava,
and in addition enter the vertebral pedicles, laminae, and
spinous processes through deficiencies in their cortex (fig.
16). The ventral half of the dura mater usually is supplied
by a small branch of a meningeal ramus (“recurrent” nerve)
which penetrates it immediately inside the intervertebral
T h e awnulus fibrosus. Nerves in this portion of the intervertebral disc are confined almost entirely to a thin lamina
of loose connective tissue at its surface, continuous with the
fibrous portion of the periosteum of the vertebral bodies. The
lamina contains small blood vessels, vasomotor nerves to them,
and a few unmyelinated, beaded axons with free endings,
which probably are sensitive to painful stimuli. These are
derived, like nerves in the nearby periosteum, from branches
of the paravertebral autonomic plexus, from recurrent nerves
to the dura mater and the posterior longitudinal ligament,
or from nerves to the anterior longitudinal ligament (figs.
2-5). No nerves have been found within the layers of the
annulus fibrosus itself, with the exception of some fibers whose
terminations occur in the posterior longitudinal ligament ;
these are restricted to the external laminae only, and are
derived from the “recurrent” nerves (figs. 7 and 17).
T h e periosteum awd bone. These structures are supplied
by numerous small branches of nearby nerves derived from
the autonomic rami and the paravertebral plexus where they
lie near the vertebral bodies, from nerves to the dura mater
and longitudinal ligaments, and in the case of the neural
arches, from nerves supplying overlying muscles (figs. 2-5).
The periosteum is supplied with fibers to the vascular
plexus, with a few beaded, unmyelinated axons ending freely,
with rare, moderate sized fibers having small complicated,
encapsulated terminations, and the periosteum conveys to
bone myelinated axons of about 5 - l o p which enter spongy
bone through deficiencies in the cortex.
The iwterspiwous a w l fEaenl ligamefints. These ligaments are
supplied externally by fibers derived from nerves to overlying muscles, and on their deep surfaces by nerves which also
innervate the dorsal part of the dura mater (fig. 16). Although
nerves accompany blood vessels into openings in the ligaments
no fibers could be found which ended within the substance of
the ligaments. These nerves apparently are destined to form
vasomotor terminations, to enter bone, or to form free terminations at the ligaments’ surfaces, in a manner similar to
that described for the annulus fibrosus.
Blood vessels. Branches of the segmental vessels which
supply the vertebral column and its muscles, ligaments, and
fasciae are innervated in a variable way. Large arteries near
the sides of the vertebrae receive fibers from the paravertebral
autonomic plexus, described previously. More peripherally
vessels entering the joints, bone and various ligaments pick
up nerves carrying not only vasomotor fibers, but others
destined to form sensory terminations in these structures.
These latter nerves occur as branches of the dorsal and ventral
primary rami, or as branches of the sympathetic trunk, where
it is nearby.
Fascia. Only the lumbodorsal fascia and the areolar fascia1
layer enclosing it have been examined. They are supplied by
continuations of nerves to the underlying sacrospinalis muscle
and therefore usually have as their source a nerve root
one level cranial. Terminations of only two types were found :
vasomotor, and small, unmyelinated axons ending in many
arborizing, tapering branches, a type which is illustrated in
figure 6, and described previously in the longitudinal ligaments.
Cervical, thoracic and lumbar diarthroses. These articulations have a fibrous capsule and are lined by synovial tissue.
Their nerve supply is typified by the knee joint, described
in detail by Gardner ( '44,'48). The basic plan of innervation
in these joints of the monkey is similar to that of the knee of
the mouse, cat and man.
The larger nerve-bundles entering the fibrous capsule arise
from several sources. Dorsal primary rami of cervical, thoracic, and lumbar spinal nerves supply two successive articulations, and each joint therefore has a bisegmental innervation. I n the lumbar region (fig. 1),a deep branch of the
dorsal ramus loops under an accessory process, turns cranially
and medially to supply the multifidus, semispinalis and rotatores muscles, and then gives off a small branch to the
dorso-caudal aspect of the capsule of the joint at the level
from which the nerve is deTived. Another, more lateral,
branch of the same dorsal ramus is directed caudally, lateral
to the accessory process, supplies parts of the sacrosginalis
muscle, and then sends a branch to the articulation a segment
caudal. Thus, a single lumbar spinal nerve supplies an articulation near its emergence as well as another joint one
vertebra caudally.
A thoracic dorsal ramus usually sends a branch under the
transverse process. This branch is directed medially and
cranially to the deep oblique muscles, then continues to the
joint one segment cranial. A caudally directed, more lateral
branch supplies the deep oblique and sacrospinalis muscles,
and in its course past the caudal joint gives off one o r two
small nerves (fig. 2).
Dorsal rami of midcervical spinal nerves generally do not
form their main divisions to muscle groups until in their
dorsal directed, slightly caudal course they lie opposite an articulation (figs. 3, 5 and 18). At this point one o r two nerves
are given off to the lateral and dorso-caudal parts of the capsule, accompanying entering blood vessels. Other branches of
the same ramus are directed caudally to an adjacent articulation. I n this respect a cervical spinal nerve resembles a
lumbar nerve in that it innervates a joint at its level of
emergence as well as one a segment caudally. Small, independent branches usually arise from the dorsal ramus near
its origin and supply not only intertransverse muscles but
also the ventral part of the capsule of a joint at the same
level or one a segment caudal.
The nerve bundles described above are less often independent “joint nerves” than they are branches of nerves
which pass through and supply skeletal muscle in tlicir course
to the joint (fig. 19, and figs. 1-5). Such a plan is typical of
nerves to the major joints of the mouse, cat and man.
Fig. 8 Entering from the left are two large, myelinated axons from the fibrous
capsule of a cervical joint. Each gives rise to a number of Ruffini terminations.
Extension of the myelin tovvard the end organ and the varicose appearance of
both the mother fiber and its terminations are typical. Redrawn from a whole
mount stained with methylene blue. X 200.
The iritracapsular course of nerves to these joints has been
followed in cleared, methylene blue stained whole mounts.
Axons entering the fibrous capsule vary in size from the
smallest autonomic fibers to those with an axis cylinder equal
in size to somatic motor axons. Axons entering the synovial
tissue and its villi appear to be limited t o vasomotor fibers.
Those terminating in the fibrous capsule vary in size and
mode of ending. The smallest usually are beaded and end in
terminal enlargements. Other small fibers which usually are
not beaded in this method of preparation dichotomize repeatedly and branch into 5 to 10 delicate arborizing endings
covering a wide area, and are similar to those found in the
longitudinal ligaments, fascia, and other fibrous connective
tissues of the body; they are illustrated in figure 6. The last
two types of fibers and endings probably subserve painful
sensations. It is known that pain can be aroused in the
intervertebral joints of man by electrical stimulation.
Large, heavily myelinated fibers in the joint capsule are
somewhat less numerous but appear prominent because of
their size and frequent division. These remain myelinated
almost to their terminations in groups of Ruffini endings (fig.
8). Nerves supplying vertebral joints are also illustrated in
figures 20 and 21.
The plexus of autonomic nerves lying lateral t o the intervertebral foramina and interconnecting the sympathetic chain
and its rami with ventral and dorsal rami of a spinal nerve
offers a direct route by which sympathetic fibers may reach
peripheral structures supplied by these spinal nerve rami
(figs. 4 and 5). I n the description of routes traversed by
sympathetic neurons the connections of the sympathetic chain
to somatic nerve plexuses and t o cranial nerves via gray and
white rami communicantes have always been emphasized. The
paravertebral plexus described here may have been overlooked only because some previous investigations have been
confined to gross dissection (Riegele, '26; Siwe, '31). This
paper intends to call the plexus to attention not only for its
possible role iii routing of autonomic fibers t o rami of spinal
nerves, but because the plexus provides paths for the distribution of somatic fibers to many of the intrinsic structures of
the vertebral column. Branches of this net are found constantly a t all spinal levels in the monkey, innervating longitudinal ligaments, spongy bone of the vertebral arches and
bodies, periosteum and the dura mater. Only structures
dorsal to the spinal nerve roots derive their nerve supply
directly from rami of the segmental nerves, and even these
communicate with the paravertebral plexus (figs. 4 and 5).
The size and content of fibers in this plexus have not been
investigated, but even 50 1.1 myelin-stained sections reveal a
great variety of fiber size. This evidence alone makes it
probable that the paravertebral nerve plexus includes more
than just pre- and postganglionic sympathetic neurons. Knowledge of its communications with both somatic arid visceral
nerves and its terminations in hone, periosteurn, and longitudinal ligaments in a variety of receptors points it out as
a system of mixed nerves. It would be interesting to know
if the plexus contains ganglion cells, to know whether its
smaller fibers are pre- or postganglionic or both, and to
analyze its nerve fiber components for size and function.
Terminations of nerves in the structures described are
limited in type. Only vasomotor, Ruffini, free simple terminations, and (in periosteum) a few encapsulated endings have
been found. All structures examined have the usual perivascular network of axons ending on arterioles, and all have
small, unmyclinated fibers, usually beaded, ending freely,
unassociatcd with vessels. Small fibers with widely-arborizing endings usually associated with reception of painful
stimuli are found in the fibrous tissues examined everywhere
but in periosteum. Only periosteum has complex encapsulated
endings, as described by other investigators, and believed to
subserve pressure sense or deep painful stimuli. Ruffini endings are limited to the diarthrodial joints and probably are the
terminal organs of neurons primarily proprioceptive in function.
I n a recent report Ikari (’54) has described the various
types of nerve terminations found in silver-stained sections
of fasciae, skin, periosteum of the vertebral bodies, annulus
fibrosus, ligaments, and vertebral articular capsules in the
human lumbar region. Nerves were found by Ikari in longitudinal ligaments and the vertebral body periosteum and
none in the annulus fibrosus, in agreement with the material
reported in this paper. However, Ikari reports finding no
nerves in fasciae and he states that there is “very little participation by nerve elements” in joint capsule. The present
report finds nerves in both of these structures, with two
distinct types of terminations in fascia and three in joint
capsule. The disparity in results may result either from different technical methods o r in dissimilarity of the monkey and
It is usually assumed that the Ruffini endings in joints serve
a reflex function, and that smaller fibers, such as have been
described, with their uncomplicated or arborizing endings,
respond to painful stimuli, Since tlie Ruffini ending is limited
to tlie diartlirodial joint capsules it seems probable that stimuli effectivein arousing postural o r locomotor reflexes emanate
only in these diarthroses. ITowever, specificity of a given
nerve termination for a limited function, such as the association of Ruffini endings with postural reflexes can only be
susptctcd. Neurophysiological experiments to test this should
be pcrforrned, and verification of segmental duplication of
the nerve supply to joints, fasciae, muscles, and longitudinal
ligaments should likewise be made.
Terminations usually held responsible for receiving painful
stimuli hare been found in thc monkey in the longitudinal liganients, periosteurn, lumbodorsal fascia, intervertebral diarthroses, and in the loose connective tissues at the surfaces of
the annulus fibrosus and ligamentum flavum. If the distribution of pain-sensitive endings is similar in man to that described here it becomes clear that pressure on a nerve root
is only one o f a number of possibilities which must bc considered in the etiology of back pain. Reflex muscle spasm arid
reference o f pain to distant areas, phenomena frequently
rioted clinically, can be explained by knowledge that the
sourccs of painful stimuli in vertebral tissues are innervated
by the same nerve roots which may contribute t o the brachial
and lunibo-sacral plexuses. Segmental overlap in nerve supply
for the spinal structures involved unfortunately may complicate accurate localization, but must nevertheless be kept
in mind.
It has been noted in this material that overlap o f adjacent
rami of the spinal nerves exists, and that branches of the
nerves supplying longitudinal ligaments overlap the midline.
Previous studies (Tl’iberg, ’49) by Luschlsa, Hovelacque,
Roofe and Wiberg disagree in the vertical extent to which
the “recui-rent” (“sinu-vertebral”) nerves ramify. All of
these studies were made by gross dissection. The present
research is based upon whole thickness preparations of tissues
supplied by these nerves and reveals a one-segment overlap of
axons conveyed in these nerves. Such an overlap of innervation in longitudinal and interspinous ligaments, joints and
muscles, if it follows a similar pattern in man, could conceivably play a part in the difficulty in accurate localization of
pain arising from these sources. Overlap of adjacent spinal
segments occurring in the deep structures of the vertebral
column is not unexpected and need not be explained as an
isolated phenomenon. The overlap of cutaneous areas supplied by these same segmental iierves has long been recognized.
An investigation of the anatomical and functional aspects
of this problem has recently been undertaken by Gardner,
Pedersen and Blunclr ( ’ 5 5 ) . They found a similar overlap
in human fetal material, and elicited reflex responses in
posterior thigh muscles on stimulation of deep lumbar structures in the cat.
Verification o f Hovelacque ’s conclusion that the recurrent
nerves may have more than one source is made here in the
monkey. However, though Hovelacque indicated both a somatic and sympathetic origin o f fibers, he failed to show that
nerves re-entering the intervertebral foramen are usually
multiple. Of course, the monkey and man may differ in this
respect. Because o f the complicated origin of these recurrent
bundles from either o r both spinal rootlets and from the paravertebral autonolmic plexus it is impossible to be certain of the
identity of these nerves when dissected grossly. Near-microscopic size and an origin near or within the foramen were
other factors complicting identification in this aspect of the
investigation. For this reason it may be difficult to proceed
with a study in which the recurrent nerve is used in a neurophysiological investigation.
The afferent limb of reflexes related t o forelimb posture
in the cat has been shown recently to arise in at least the
upper three cervical diarthroses. Responses from cervical
muscles in the same animals were shown to have an insignificant effect on alterations in forelimb position (McCouch,
Deering and Ling, ’51). After labyrinthectomy and resection
of cervical niuscles in acute experiments, the forelimb response to head rotation was retained, but it was abolished
after cutting around the dorsal rami of the spinal nerves so
as to denervate the diarthroses. I n some chronic animals
this response of the forelimbs returned in an “inconstant,
evariesccnt form.” If in the cat deep nerves supplying the
ventral portion of the joint arise as they do in the monkey,
by traversing intertransversarii muscles (fig. 3), retention
of the reflex can be explained by assuming that this nerve
was not severed merely by circumcising the dorsal ramus.
Ruffini endings are present in cei-vical joint capsules of the
cat (Catalano, ’55), a fact consistent with a reflex function
existing here.
The origins and terminations of nerves to structures of the
vertebral column and its associated structures in the monkey
have been described. I n large part these nerves are branches
of a segmentally-arranged paravertebral plexus which communicates with gray and white sympathetic rami, with spinal
and sympathetic ganglia, and with the branches of the dorsal
and ventral spinal nerve rami. These communications convey
axons of various size and degree of myelination, and appear
to comprise a network of mixed nerves. Somatic sensory
and autonomic nerves are given off the plexus, to end in
longitudinal ligaments, blood vessels, bone marrow, periosteum, joints and dura mater.
A one-segment overlap in the innervation of vertebral
structures has been demonstrated, as well as for the dorsal
deep muscles, corresponding to a similar cutaneous dermatome overlap. Such an arrangement may complicate accurate
clinical localization of painful sensation arising from these
structures, since a similar situation exists in man. Nerve
terminations usually considered pain-sensitive have been
found in the longitudinal ligaments, periosteum of the vertebral bodies, in loose connective tissues at the surfaces of the
annulus fibrosus and ligamentum flavum, in the lumbodorsal
fascia, and in the fibrous capsules of the vertebral diarthroses,
emphasizing the numerous possible sources of pain from
spinal structures.
The meningeal ramus (recurrent o r sinu-vertebral nerve)
is both somatic and autonomic in content, is at least two in
number to a segment, and supplies sensory and vasomotor
nerves to dura mater, periosteum of the spinal canal, and
the posterior longitudinal ligament, as well as to spongy bone
of the vertebral bodies and arches.
Dorsal rami of the spinal nerves supply deep dorsal muscles,
periosteum and bone of the vertebral arch, and provide several sensory branches to the intervertebral diarthrodial joints.
Each joint receives nerve fibers from two adjacent spinal
nerves, entering from a variety of directions, and varying
slightly in pattern with the vertebral level. Nerves and their
terminations in these joints follow a plan similar to those
described in the major joints of other mammals.
CATALANO,J. V. 1955 Personal communication.
1947 A new method f o r
investigating the nervous system. Brain, 70 : 495-506.
GARDNER,E. D. 1943 Surgical anatomy of the external carotid plexus. Arch.
Surg., 4 6 : 238-244.
1948 The innervation of the knee joint. Anat. Rec., 101: 109-130.
1944 The distribution and termination of nerves in the knee joint
of the cat. J. Comp. Neur., 80: 11-32.
1948 The nerve supply of diarthrodial joints. Stanford Med. Bulletin, 6, No. 3, 367-373.
AND C. 1”. J. BLUNCK1955 The anatomy of
lurnbosacral posterior primary divisions and sinu-vertebral nerves, with
a n experimental study of their function. Anat. Rec., 122, No. 2 : 297.
A. 1925 Le nerf sinu-verthbral. Ann. d’Anat. Path., 9 : 4 3 5 4 4 3 .
CHU 1954 A study of the mechanism of low back paiu (abstract). J.
Bone and Jt. Surg., 3 6 8 : 195.
A,, AND A. BRUNSCHWIG1932 Recherches Histologiques sur l’innervation
des articulations des corp vertbbraux. Presse MBd., 40: 316-317.
KUNTZ,A., AND C. A. RICHINS 1945 Innervation of the bone marrow. J. Comp.
Neur., 83: 213-223.
1936 The innervation of the pcriodontal
membrane. J. Anat. (London), ’71: 98-102.
McCoucr~,G. P., I. D. DEERIXGA N D T. H. LING 1952 Location of receptors f o r
tonic neck reflexes. J. Kenrophysiol., 24 : 191-195.
MCMANUS,J. F. A. 1946 The demonstration of certain f a t t y substances i n
paraffin sections. J. Path. Bact., 58: 93-95.
RIEGELE,L. 1926 Uber der Innervation der Halti- und Brustorgane bei einigen
Affen. Zeit. f. Anat. u. Entw., 80: 777-858.
ROOFE, P. 1940 Innervation of annulus fibrosus and posterior longitudinal ligament. Arch. Neurol. and Psych., 4 4 : 100-103.
SFARIENI, A. Rccherches anatomiqucs sur l’existence des nerfs et sur leur mode
cle se terminer dans le tissu adipeux, dans le pbrioste, dans le pCrichoiidre
et dans les tissus qui renforceiit les articulations. Arch. Ital. 3e Riol.,
38: 49-101, 102.
SIWE,S. A. 1931 The cervical p a r t of the ganglionated cord, with special reference to its connections with the spinal nerves and certain cerebral
nerves. Am. J. Anat., 48: 479-497.
WIBERG,G. 1949 Back pain in relation to the nerve supply of the intervertebral
disc. Acta Orth. Scan., 1 9 : 211-221.
Note: Prior to publication of this paper a study of the innervation
of man’s vertebral column has appeared, “The anatomy of lumbosacral
posterior raini and meningeal branches of spinal nerves (sinu-vertebra1 nerves).” J . B. and Jt. Surg. 38A: 377-391, 19.56. I€. E. Pedersen, C. F. J. Blunck and E. Gardner have described the origin and
distribution of the sinu-vertebral nerve (meningeal ramus) and
branches of the posterior rami which supply spinal structures. Using
different technical methods, their anatomical findings are similar to
those reported here. Stimulation of deep back structures in cats produced local and distant reflex mnscular responses, suggesting mechanisms by which referred pain and mnscle spasm may occur in man.
Meningeal ramus (“recurrent nerve”), indicated by arrows, lying adjacent to the scgmeiital vein at a
mid-thoracic level. I t s destinations are the posterior longitudinal ligament (P.L.L.) and the ventral
portion of the dura mater ( D ) . R., dorsal rami of the spinal nerve; S.C., spinal cord.
Nerve in the posterior longitudinal ligament. Division of a small axon is shown. Methyleiie blue. 150 X.
Mid-lumbar transverse section. The vertebral body, anterior loiigitudiiial ligament (A.L.L.) , sympathetic
trunk (S), and psoas major muscle (P) a r e labelled. The arrow ilidicates a branch of the sympathetic
trunk which innervates the anterior longitudinal ligament and supplies fibers to marrow and vessels
of the vertebral body. Sudan black. 15 X.
Sagittal servical section through the vertebral plexus. Ranius eommunicaiis (R.C.) contributing niydinated
fiber8 to a trunk of the vertebral plexus (V.Pl.). There is variability in fiber size and in myelination.
Sudan black. 50 X.
Transverse section through C - 5 . This shows the relations of the vertebral vessels ( A and V ) , the spinal
ganglioii (S.G.), and three rami communicaiites (arrows) contributing to the autonomic plexus ( P ) , which
enmcsiies the artery. Sudan black stain. 15 x.
Transverse lumbar section. Spinal cord, 8.C.; spinal ganglion, S.G.; dura mater, D.; accessory process,
Ace. P.; 4 branches of the dorsal ramus of the spinal nerve, r.; autonomic ramus communicans, R.C. Among
and around the dorsal ramus branches are a number of the nerves comprising the paravertebral plcxus.
A branch of the paravertebral plexus lies dorsal t o the spinal ganglion arid carries nerves t o the dorsal
half of the dura mater and to the vessels entering the neural arch. (See arrows). Sudan black.
30 X.
Tangential section at the lateral edge of the annulus fibrosus, thoracic. A number of mpelinated wrves
(arrows) accompany small blood vessels. Sudan black. 100 X. I n metliylene blue preparations these iiervvs
are seen to end on the vessels in nearby periosteum, and a few to penetrate into the vertebral body. There
are no terminations within the disc itself. (See text.)
14 and 15 Nerve terminations in the posterior longitudinal ligament. Figure 14 is a termination similar
to one illustrated by the drawing in figure 6. Because of the extent of their raniification these are
difficult to photograph satisfactorily. I n figure 15 the ending of a small, unmyelinated fiber in a grapelike cluster near the surface of the ligament is seen. hfethylene blue. 150 X.
18 Tangential sagittal section through the lateral portion of two mid-cervical
joints. The dorsal ranius (D. R.) is seen at the right before it passes
lateral to the upper joint, and at the left, after it has divided into 4
branches (numbered). The upper three are destined for deep dorsal muscles,
and the lowrst (4) to the joint, fascia, and deep muscles a segment caudal
to its origin. Arrows indicate branches of this dorsal ramus which penetrate
and supply the joint capsule. Other linear structures resembling nerves are
blood vessels. Sudan black. 40 X.
Horizontal section of a lumbar joint. Nerves pointed out bv the arrows
accompany blood vessels to the region just caudal to the fibrous joint capsules, prior to penetrating it. The position of the capsule is indicated by
the label. Sudan black. 25 X.
Cleared, full-thickness section of the lateral portion of a cervical joint
capsule, stained by intravital methylene blue. Below is a p a r t of the
dorsal ramus; a sniall branch arises from its upper surface, containing perhaps 8 axon8 which vary in size, and are bound for the nearby fibrous joint
capsule. 150 x.
Sudan black stain of a bifurcating myelinated nerve i n a thoracic joint capsule. Variation i n fiber size and myelination are evident. 150 X.
Без категории
Размер файла
1 742 Кб
structure, associates, monkey, nerve, vertebrate, supply, column
Пожаловаться на содержимое документа