Gardner 1948

Download as pdf or txt
Download as pdf or txt
You are on page 1of 22

T H E INNERVATION O F THE KNEE JOINT1

ERNEST GAR,DNER
Department of Anatomy, College of Medicine, Wayne University,
Detroit, Michigan

FOUR FIGURES

INTRODUCTION

The probable role of articular nerve supply in many physio-


logical mechanisms has been discussed previously (Gardner,
'48). It is obvious that an understanding of the functions of
this nerve supply depends to a large extent upon a much more
precise knowledge of its anatomy than now exists. The more
important anatomical features which should be considered
include gross and microscopic distribution of the nerve fibers,
types of endings and their relationships t o various layers
and regions of joint capsules, fiber diameters in articular
nerves, dorsal root inflow and central nervous system path-
ways, This and succeeding papers deal mainly with the gross
and microscopic distribution of articular nerves in man.
There have been a number of studies of the distribution of
nerves to various diarthrodial joints, but most of these were
done during the last century and there has been a surprisingly
incomplete or inaccurate transference of such data to modern
textbooks of human anatomy. There have been no studies of
the intra-articular distribution of nerves. Information per-
taining to this is of value since it allows a comparison with
what is known regarding other animals. F o r example, it has
been shown that in the knee joint of the mouse (Gardner, '42)
and of the cat (Gardner, '44) there are regional differences in
distribution and in types of nerve endings.
'Aided by a grant from the Division of Research Grants and Fellowships,
National Institute of Health, U. 8. Public Health Service.
109
110 ERNEST QARDNER

Ellis in 1840 described most of the nerves to the human


knee joint and specifically mentioned 5 branches in the pop-
liteal fossa. One of his illustrations in a later publication
(El&, 1882) shows a branch of the nerve to the vastus
medialis ramifying in the medial part of the joint capsule
as far as the patellar ligament.
Cruveilhier (1844) also described the nerves to the knee
joint. He pointed out that the articular branch of the nerve
to the vastus lateralis not only perforates the thick, fibrous
layer of the capsule, but also extends to and ramifies in the
infrapatellar fat pad. I n addition, he stated that the
saphenous nerve, while in the sheath of the adductor magnus,
gives an articular filament which descends vertically, per-
forates the capsule and enters the synovial adipose tissue. He
also mentioned 5 articular branches of the common peroneal
and tibial nerves, corresponding to the genicular arteries.
Riidinger (1857) appears to have been the first to make a
systematic study of the nerves to various joints. For the knee
joint he described branches from the tibial, common peroneal
and saphenous nerves and from the nerves supplying the
vastus medialis and lateralis muscles. He stated that those
from the tibial and common peroneal nerves formed an ex-
tensive popliteal plexus which supplied the back of the knee
joint. He did not mention the obturator nerve, nor did he give
any indication of the number of dissections on which his work
was based. I n addition t o his own investigations, he reviewed
the literature concerning nerves to bones, ligaments and
joints.
Sappey (1877) mentioned articular branches of the tibial
nerve, and of the nerve to the vastus medialis. Fick ('04)
stated that the articular nerves correspond exactly with the
vessels to the joint. I n Quain's PeripheTal Nerves and Sense
Organs ('09), the branches of the nerves to the vasti are
described, as well as 5 genicular nerves arising from the
tibial and common peroneal nerves. No mention was made of
a possible articular twig from the saphenous nerve.
NERVE SUPPLY OF KNEE JOINT 111

Druner ('27) studied the role of the obturator nerve in the


innervation of the knee joint. He pointed out that the so-called
cutaneous branch which lies in the adductor canal gives a
filament to the anteromedial part of the joint. I n addition,
he found branches from the posterior division which reached
the back of the joint capsule.
The accounts in current textbooks of human anatomy are
brief and sometimes inaccurate. For example, Piersol's
Human Anatomy ('30) states that the articular branches of
the tibial nerve are small and variable in number and break
up into fine filaments which inosculate with the lower articular
fibers of the common peroneal and form the popliteal plexus of
Rudinger. This description is similar to that of Rudinger
(1857) who had overlooked the obturator nerve in this regard.
As will be pointed out later, the common peroneal nerve
rarely if ever enters into this plexus. Piersol also mentions
an articular branch of the saphenous nerve which is said to
be inconstant.
Morris' Human Anatomy ('42) mentions twigs to the knee
joint from the nerves to the vastus medialis, intermedius and
lateralis, a branch from the posterior division of the obturator
nerve, superior medial, middle (azygos) and inferior medial
genicular branches from the tibial nerve and superior and
inferior lateral genicular branches from the common peroneal
nerve. No mention is made of a possible articular branch of
the saphenous nerve.
The account in Cunningham's Text-Book of Anatomy ('43)
is somewhat more complete. The femoral nerve is described
as giving 4 branches to the knee joint: 1 each from the nerves
to the vastus lateralis, intermedius and medialis, and some-
times 1 from the saphenous nerve. The continuation of the
posterior division of the obturator nerve reaches the knee
and pierces the oblique popliteal ligament. A branch to the
anterolateral side of the joint arises either from the common
peroneal nerve, or from the sciatic nerve in common with the
branch to the short head of the biceps femoris, and divides
112 ERNEST BARDNER

into branches which accompany the lateral genicular arteries.


The recurrent peroneal nerve also has an articular branch.
There are usually 2 branches of the tibial nerve, one piercing
the oblique popliteal ligament, the other accompanying the
inferior medial genicular artery. The account in Gray's
Anatomy of the Human Body ( '42) is essentially the same ; 3
articular branches of the tibial nerve are mentioned.
It is apparent that most of the nerves to the knee joint have
been described at one time or another, and that the accounts
in current textbooks have become simplified, sometimes to the
point of inaccuracy. Differences between the various original
accounts seem most easily explained on the basis of variation,
and this is borne out by the findings of Jeletsky ( '30). His
is probably the most extensive investigation of the gross nerve
supply of the human knee joint. In 30 dissections he found a
wide variation in the distribution of the various articular
nerves. He did, however, succeed in establishing the con-
stancy of certain branches and showed a definite pattern in
the innervation of the joint. His findings will be referred to
throughout this paper.
On rare occasions the accessory obturator nerve may supply
the knee joint. At least 2 cases have been recorded (Paterson,
1894; Jamieson, '03) in which this nerve had an extensive
distribution to the skin and muscles of the medial side of the
thigh and, in addition, sent branches to the hip and knee joints.
It is to be expected that in approaching a particular struc-
ture nerves will vary in the direction and pattern of their
distribution. What is more important, perhaps, is the dis-
tribution and termination of nerves after they have entered
the structure. So f a r as could be determined, no information
on this score for the human knee joint exists in the literature.
There have been a few studies of nerves and nerve endings
in human joint capsules. These were rarely if ever based on
serial sections and they provide no information regarding
regional differences in nerve supply.
NERVE SUPPLY O F KNEE JOINT 113
MATERIAL8 AND METHODS
Nerves were dissected in 11adult knee joints so as to estab-
lish the general pattern of distribution. Four other joints
were partially dissected to confirm points brought out by
studies of fetal joints. Five fetal joints were used to establish
the distribution of nerves within the joint capsule :all of them
were serially sectioned at 10 microns.2 Two were from a
9-week fetus. The right was sectioned in the frontal plane
and stained by a modified Masson’s method; the left was
sectioned in the sagittal plane and stained with hematoxylin
and eosin. The right knee joint from a 104-week fetus was
sectioned frontally and stained with hematoxylin and eosin.
Two joints from a 12-week fetus were sectioned sagittally;
the right one was stained with hematoxylin and eosin and the
left one by a modified Masson’s method. These ages are only
approximate since they were determined on the basis of
crown-rump measurements of fixed specimens.
It was possible in the fetal sections to trace even very small
nerves for long distances, since their sheaths always stained
well, and most of the nerves contained at least a few relatively
large fibers which were easily distinguished. The material,
however, was not satisfactory for the study of nerve endings.
The sections included approximately half of the thigh and
leg so that it was possible to obtain an accurate picture of
the origin, course and distribution of all the nerves to the
knee joint.
Diagrammatic sketches were made of representative dissec-
tions, and projection tracings were made of fetal sections
which showed characteristic features of the nerve supply.
OBSERVATIONS
The nerve supply of the knee joints studied in this series
was derived from the femoral, obturator and sciatic (tibia1
and common peroneal) nerves.
a The first 10 dissections were done while the author was in the Department of
Anatomy, Stanford University, California, 1940-41. The fetal material was
obtained through the courtesy of Professor C. H. Danforth, Department of
Anatomy, Stanford University, California.
114 ERNEST GARDNER

The articular branches of the fernoral nerve


These branches arise from the saphenous nerve and from
the nerves to the vastus medialis, intermedius and lateralis
muscles. In the illustrations, therefore, the articular branches
are named according to their most specific origin. Thus, “to
vastus medialis” (fig. la, ant.) indicates the articular branch
of the nerve to the vastus medialis. The same method is used
in labeling the drawings of serial sections of fetal joints.
Sapheinous inerve. When the first 10 dissections were done,
it was not realized that the saphenous nerve had an articular
division. According to Jeletsky ( ’30) this is a constant branch
which frequently includes fibers from the obturator nerve.
It was found in animals which were studied since this work
was begun (Gardner, ’48). It was present in all the fetal
joints, and its existence confirmed in each of 5 adult knee
joints. I t s level of origin is variable. It arises either within
the femoral trigone, in which case it descends on the femoral
artery, o r else it may arise just before the femoral artery
passes through the adductor magnus muscle. I n either case
it is a small branch which descends vertically, anterior t o the
tendon of the adductor magnus, and then turns toward the
anteromedial part of the joint capsule (fig. 2f, ant.). Here it
ramifies in the fibrous layer of the capsule, and some filaments
can be traced almost as far as the ligamentum patellae where
they pierce the tissue in the direction of the infrapatellar fat
pad. I n the fetal joints, this branch has a similar course
(fig. 3 and fig. 4, e-f). It anastomoses in the medial part of
the capsule with a branch from the nerve to the vastus
medialis, as well as with a bundle of fibers accompanying the
superior medial genicular artery derived from the obturator
nerve. Some of its fibers reach the infrapatellar fat pad and
ramify amongst the vascular network of this region (fig. 3d
and fig. 4e-f). As Driiner ( ’27) and Jeletsky ( ’30) pointed
out, there is frequently an anastomosis between the obturator
and the saphenous nerves in the adductor canal. It is there-
fore difficult to be certain about the composition of the
NERVE SUPPLY OF KNEE JOINT 115

Tovast. med.d /k\f /

Pop. plex.

Fig. 1 Anterior and posterior views of different types of distribution of nerves


to the knee joint. The joint capsule is represented by stippling.
116 ERNEST OARDNER

Fig. 2 Further patterns of distribution. f, ant. shows the articular branch


of the saphenous nerve.
NERVE SUPPLY OF KNEE JOINT 117

articular branch of the saphenous nerve. Jeletsky also men-


tions cases in which fibers from the saphenous nerve accom-
pany the femoral artery into the popliteal space. I n such
instances, it is quite possible that the fibers arise in part o r
entirely from the obturator nerve. I n one of the cases of the
present series, the articular branch of the saphenous nerve
anastornosed with the anterior division of the obturator nerve
and most of the branch resulting from this union accompanied
the femoral artery into the popliteal space and thence to the
back of the knee joint. In another instance, the articular
branch of the saphenous anastomosed with the anterior divi-
sion of the obturator nerve near the origin of the branch
from the femoral nerve to the pectineus muscle. In the cat,
the articular branch of the saphenous is occasionally partially
or completely replaced by a branch from the anterior division
of the obturator nerve (Qardner, '48).
Only one of the drawings in figures 1and 2 shows the articu-
lar branch of the saphenous nerve. The other dissections
illustrated were done before the existence of this was realized.
Because of its subsequent discovery in fetal and adult mate-
rial, and because of its presence in the large series of Jeletsky,
there seems to be little doubt that it is a relatively constant
branch.
The werue t o the umtus medialis. This descends along the
medial edge of the muscle, in company with the saphenous
nerve. The most distal of its branches turns around the edge
of the muscle and enters the medial portion of the joint cap-
sule, somewhat more superiorly than does the articular branch
of the saphenous. It divides into a number of twigs, some of
which can be traced to the medial edge of the patella, and
occasionally as f a r as the attachment of the capsule to the
medial tibia1 condyle. The variations in distribution are
shown in figures 1and 2. I n all the fetal joints, this nerve had
a similar course (fig. 3 and fig. 4e-f), and was clearly seen to
anastomose with the articular branch of the saphenous and
also with fibers from the obturator nerve which accompanied
the superior medial genicular artery. This anastomosis occurs
118 ERNEST GARDNER

emor'' ''
I Saphenous
yast. rned.
,

fat pad

. med. gen.

Fig.3 Projection tracings of a fetal joint sectioned in the sagittal plane.


Nerves are shown in solid black. The sections start from the medial side of
.
the joint
NERVE SUPPLY OF KNEE JOINT 119

Peroneal a vast. I$ cus

Peron
Recurrent peroneol

Fig. 4 Continuation of the sections started in figure 3.


120 ERNEST GARDRER

as the fibers traverse the fibrous layer of the capsule. Some


of the blood vessels which these nerves accompany give
branches t o the medial femoral condyle. In general, the nerves
continue to the medial edge of the patella where they ramify
in the periosteum and also accompany vessels into the patella.
Some fibers course inferiorly, penetrate the capsule just below
the patella and enter the infrapatellar fat pad where they
ramify extensively (fig. 3d and fig. 4e-f). Some were traced
as far inferiorly as the tibia1 tuberosity. The participation
of these 3 nerves in the innervation of the anteromedial por-
tion of the capsule confirms the findings of Jeletsky ( '30).
T h e nerves to the vastus intermedius. These give branches
which descend on the anterior surface of the femur and supply
its periosteum as far inferiorly as the edge of the articular
cartilage. Some twigs enter the tissue lining the suprapatellar
extension of the joint cavity (figs. 1 and 2), while others oc-
casionally reach the superomedial part of the capsule (fig. 2e,
ant.). Anastomoses with other muscular branches while
descending in the quadriceps femoris muscle are not uncom-
mon. No further features were evident in the fetal joints, ex-
cept that when the fibers reached the suprapatellar recess
(fig. 3d and fig. 4f) they always became associated with the
blood vessels supplying the neighboring portion of the femur
and also the tissue lining the recess. The findings were sim-
ilar t o those of Jeletsky ('30).
The n,erue to the vastus lateralis. The most distal of a num-
ber of branches from this nerve turns around the posterior
edge of the muscle and divides into a varying number of
filaments which enter the anterolateral portion of the joint
capsule. Here they were often traced as far as the lateral
edge of the patella (figs. 1 and 2). The level to which the
twigs descend and the extent to which they branch varies in-
versely according to the distribution of the articular branch
of the common peroneal nerve. I n the f eta1 joints, this portion
of .the capsule is supplied by these same nerves (fig. 4i-1).
During the initial part of their course both of these nerves
ramify in the fibrous layer of the capsule. Some fibers reach
NERVE SUPPLY OF KNEE JOINT 121

the periosteum of the lateral edge of the patella. Others be-


come associated with the blood vessels supplying the lateral
femoral condyle. Still others gradually penetrate the capsule
as they travel toward the inferior edge of the patella. These
reach the infrapatellar fat pad and ramify in the vascular
network here; some fibers reaching almost to the tibial
tuberosity. Jeletsky also found that the anterolateral portion
of the capsule was supplied by the common peroneal nerve
and the nerve to the vastus lateralis, although as in the cases
of the other articular nerves, he did not, by dissection, trace
the terminal fibers to the infrapatellar fat pad.

T h e articdar branches of the tibial nerve


I n many cases, the articular branches to the posterior part
of the knee joint arise from the sciatic nerve. In such cases,
however, they are definitely from either the common peroneal
or tibial portions, so they will be described separately under
these nerves.
Current textbook descriptions of the articular branches of
the tibial nerve fail to give any idea of the variations. The
standard descriptions of 3 genicular branches, corresponding
to the superior medial, middle (azygos) and inferior medial
genicular arteries were not confirmed. Usually there is but
a single large branch, the level of origin of which is extremely
variable (figs. 1 and 2 ) . I n many cases it arises in the thigh
from the tibial portion of the sciatic nerve, and in its descent
is closely bound to the sciatic nerve by connective tissue. At
other times it arises from the tibial nerve in the popliteal
fossa. I n all cases it has a variable number of branches which,
with similar branches from the obturator nerve, form a rather
dense plexus. The extent of this plexus is shown in figure 2f,
post. A similar complexity obtains for the other joints, but
terminal ramifications were omitted from the drawings. Some
twigs from this plexus enter the adventitia of the popliteal
vessels. Others accompany the genicular vessels for a short
distance, both laterally and medially. Still others, these being
122 ERNEST GARDNER

in the majority, enter the thick, fibrous layer of the joint


capsule, particularly that constituting the oblique popliteal
ligament. I n 3 of the gross dissections, a fine twig accompanied
the superior medial genicular artery for a considerable dis-
tance and thereby formed the corresponding nerve (fig. 2d,
post.). The direct continuation of the tibial articular nerve
joins the inferior medial genicular artery, accompanies it deep
to the tibial collateral ligament, and can be traced toward the
anterior part of the capsule in the region of the ligamentum
patellae (figs, 1and 2). During its course it gives off a number
of twigs, especially to the periosteum of the tibia, so that
the terminal part of the nerve becomes exceedingly small.
Evidently the plexus formed posteriorly corresponds to
the popliteal plexus of Rudinger, although it is formed by the
tibial and obturator nerves, rather than the tibial and
peroneal nerves, as he described it. Sometimes 2 separate
branches arise from the tibial nerve, and occasionally more.
According to Jeletsky ( '30), as many as 5 articular branches
of the tibial nerve may exist. I n this series of dissections, a
single branch was the usual finding.
A similar extensive distribution was seen in the fetal joints
(fig. 3 and fig. 4e-i). The fibers which enter posteriorly spread
medially and laterally in the fibrous layer of the capsule.
Some accompany blood vessels anteriorly, along the capsular
attachments to the menisci, and almost reach the infrapatellar
fat pad. It was observed that these vessels, and a few nerve
fibers, supplied the external part of the menisci and occa-
sionally penetrated deeply into their substance. Other bundles
of nerve fibers remain in the posterior part of the capsule.
They ramify in the thickened, dense connective tissue of this
region. Many have no definite association with blood vessels ;
they simply decrease in size and disappear. Still other bundles
of fibers, together with blood vessels, pierce the capsule and
course in the synovial tissue along the cruciate ligaments;
some reaching as far as the infrapatellar fat pad (fig. 3d and
fig. 4e-f) and others remaining with the blood vessels which
supply the ligaments and the nearby portions of the tibia and
NERVE SUPPLY OF KNEE JOINT 123

femur. The inferior medial genicular nerve is prominent in


the fetal joints just as it is in the adult (fig. 3 and fig. 4e-g).
It curves around the knee joint and is distributed mainly to
the periosteum, but also to the vessels supplying the medial
tibial condyle. The few terminal fibers can be traced almost
to the lower part of the ligamentum patellae.
Just below the popliteal space a few twigs from the tibial
nerve are given to the periosteum of the fibula, and some can
occasionally be traced to the superior tibio-fibular articula-
tion and to the most inferior portion of the capsule of the
knee joint.

T h e articular bralzches of the common and recurrent


peroneal nerves
The comrno!n peroneal !nerve. In one of the dissections, no
articular branch of the common peroneal nerve was found.
I n this particular instance the articular branch from the nerve
to the vastus lateralis had an extensive distribution to the
anterolateral portion of the joint capsule (fig. l a , ant.). I n
the remaining cases only a single articular branch was pres-
ent. This, like the branch of the tibial nerve, is variable in
its level of origin. Frequently it arises from the peroneal
portion of the sciatic nerve about halfway up the thigh. At
other times it is given off in the popliteal fossa. Whatever
the level of its origin, the branch joins either the superior
or inferior lateral genicular artery or else gives twigs to both
(fig. 2d, post.). When the superior branch is present, it
accompanies the artery deep to the tendon of the biceps
femoris muscle and reaches the anterolateral portion of the
joint capsule. The extent of its distribution is inversely pro-
portional to that of the articular branch from the nerve to
the vastus lateralis muscle which reaches approximately the
same area. Occasionally, fine filaments can be traced as far
inferiorly as the edge of the lateral tibial condyle, and ante-
riorly almost as far as the ligamentum patellae. In the fetal
joints, the distribution of the superior branch has been
124 ERNEST GARDNER

described in connection with the articular branch from the


nerve to the vastus lateralis (fig. 4g-1).
When fibers accompany the inferior lateral genicular
artery, they ramify in the lateral portion of the capsule as
far inferiorly as the lateral tibia1 condyle (fig. 4g-1) and
course with the branches of the vessels supplying the capsule
and the condyle.
The recurrefit perorcelal vwrue. In all the dissections,
branches from this nerve were followed to the periosteum of
the anterolateral surface of the tibia, almost to the level of
the capsular attachment. They became too small to be
dissected much farther. In the fetal joints, these fibers ac-
company blood vessels which supply the anterolateral portion
of the tibia and some continue superiorly, pierce the capsule
of the knee joint and enter the infrapatellar fat pad (fig. 4h-j).
The recurrent peroneal usually gives twigs to the superior
tibio-fibular joint as well.

The articular brarcches o f the obturator merue


In 2 cases, no branch of the posterior division of the obtura-
tor nerve to the knee joint was found. This does not mean,
however, that fibers from the obturator nerve may not have
reached the joint for, as pointed out previously, the frequent
anastomoses of the anterior division with the articular branch
of the saphenous nerve make this quite probable. I n one case,
fibers resulting from such a union traversed the adductor
canal and accompanied the popliteal vessels to the knee joint.
According to Driiner ('27) the anterior division of the obtura-
tor regularly gives a branch to the knee joint.
In the rest of the dissections the posterior division con-
tinued through the adductor magnus and reached the popliteal
fossa by descending on the popliteal vessels. I n one instance
it branched extensively and formed a plexus around the
popliteal artery (fig. 2e, post.). Usually it gives at least a
few filaments to the artery. Its subsequent distribution is
variable. I n 4 of the dissections, and in all of the fetal joints,
small twigs were given to the superior medial genicular artery
NERVE SUPPLY OF ENEE JOINT 125

which they accompanied to the same region of the capsule


supplied by the saphenous nerve and the nerve to the vastus
medialis. The direct continuation of the posterior division
breaks up into fine filaments which anastomose with similar
branches of the tibial nerve and form the popliteal plexus.
It was determined by the study of the fetal material that these
obturator fibers supplied mainly the superior part of the
posteromedial capsule and ramified mainly in the fibrous layer
of the capsule (fig. 3 and fig. 4e-f).

DISCUSSION

It is apparent that there is a basic pattern of distribution


of nerves to the knee joint and that in spite of considerable
individual variation the pattern is similar to that found in
other animals which have been studied in detail, that is, the
mouse and cat. Branches of the tibial and saphenous nerves
to the knee joint of the horse have been described (Ghetie,
'39). Taylor ( '43) also refers to nerves supplying the knee
joint of the frog. He mentions 3 branches appearing early
in the development of the limb bud which nevertheless are so
small in the adult that only 1 has been described in the
literature. This is the ramus articularis genu et pedis (Ecker,
1896) which leaves the peroneus near the knee joint. Of the
other 2, which Taylor suggests are proprioceptive, one runs
from the peroneus to the anterolateral surface of the joint
and the other from the tibialis profundis to its median region.
These Taylor suggests designating rami articulares genu
lateralis and medialis respectively.
It is apparent that the relationship expressed by Hilton
(1891) hold here. All of the nerves which supply muscles
bringing about a major movement at the knee send branches
to the knee joint. The real significance of this statement,
known as Hilton's law, must await the experimental demon-
stration of some correlation between afferent impulses arising
in the joint and changes in muscular activity resulting there-
from.
126 ERNEST GARDNER

Some of the nerves to the joint are direct branches, that


is, they do not arise from muscle nerves, but the significance
of this, if any, is unknown.
Individual variation in the nerve supply of the knee joint
is expressed in the manner in which the nerves reach the joint
and also by the fact that some of them are occasionally absent.
F o r instance, the articular branch of the posterior division of
the obturator nerve is sometimes absent, at least as far as
this series is concerned. When present, it may or may not
give rise to a grossly recognizable superior medial genicular
nerve. In view of the fact that the articular branch of the
tibia1 nerve is so commonly single, it hardly seems worthwhile
to describe separate superior, middle and inferior genicular
nerves according to the manner of most of the current text-
books. Likewise, the articular branch of the common peroneal
nerve, when present, is commonly single. I n most respects,
this series of dissections corroborates the results obtained by
Jeletsky. Since his series is by far the larger, it is presumed
to be more accurate as to the numbers and types of variations.
Once the nerves enter the capsule their course is more
constant. The majority of the bundles of fibers are closely
associated with blood vessels, just as is the case in the knee
joints of the mouse and cat. The fibers lie next to or within
the adventitia of the vessels and decrease in size as the vessels
decrease. This almost certainly indicates terminations in
association with the vessels. These vessels, incidentally, sup-
ply the neighboring epiphyses as well as the joint capsule. It
was also observed that any nerve bundles which entered the
synovial layer always did so in company with blood vessels.
The more dense, fibrous layer of the posterior capsule had
the heaviest concentration of nerves. These broke up into
numerous small bundles, many of which had no direct associ-
ation with blood vessels. If the knee joint of man is similar
to that of other animals, insofar as nerves are concerned, it
may well be that proprioceptive endings occur in this region.
The overlap of nerves within the joint is quite great. For
example, the fibers in the anteromedial portion of the capsule
NERVE SUPPLY OF KNEE J O I N T 127

and in the infrapatellar fat pad are derived from the nerves
to the vastus medialis, and from the obturator, the saphenous
and the tibial nerves. A similar region of the lateral side is
supplied by the nerve to the vastus lateralis, and the tibial,
common peroneal and recurrent peroneal nerves. The poste-
rior portion of the capsule is supplied by the obturator
(posterior division) and tibial nerves and possibly by the
saphenous and anterior division of the obturator. No nerve
supplies a portion of the capsule which is not reached by
another nerve.
When the number of nerves reaching the joint is considered,
the segmental representation must be extensive. According
to Paterson (1894) the root value is the third lumbar to the
first sacral segments inclusive, with the fourth lumbar having
the greatest representation, and then the fifth lumbar. I n
postfixed plexuses, it is the fifth lumbar and then the first
sacral segments. I n the cat also, myelinated fibers from the
knee joint enter the spinal cord over 3 and sometimes 4 dorsal
roots (Gardner, ’48).
There is little direct evidence bearing on the functions of
these nerves. There is, of course, no doubt that some of the
fibers carry impulses resulting from painful stimuli, but the
experimental evidence concerning the origin of the pain is
controversial. For example, according to Lennander ( ’06)
and Raszeja and Billewicz-Stankiewicz ( ’34), synovial mem-
brane is directly sensitive to painful stimuli. Leriche (’30)
and Brunschwig and Jung ( ’32), on the other hand, pointed
out that in cases of severe sprain, the intra-articular injec-
tion of a local anesthetic does not relieve the pain, whereas
a peri-articular injection does. Nystrom (’17) on the basis
of clinical studies concluded that periosteum and joint cap-
sules were sensitive t o painful stimuli, but cartilages and
“joint surfaces” were not.
There is abundant clinical evidence that painful joints are
often accompanied by reflex muscular contractions. There
is, however, no direct proof that this results from stimulation
of endings in the synovial tissue of the joint, nor has there
128 ERNEST GARDNER

been any experimental study of the relationships between


specific muscle groups and specific articular nerves.
I n view of the association between blood vessels and many
of the fibers in articular nerves, one is justified in assuming
that these fibers are vasomotor and vasosensory, but again
there has been no experimental work bearing on this point.
It is not unlikely that proprioceptive endings exist in the
knee joint but their exact functions are unknown. Even their
role in position sense is uncertain since in clinical methods of
testing it is impossible to dissociate stimulation of tactile,
pressure, muscle and tendon endings.

SUMMARY
The nerve supply of the human knee joint was studied in
dissections of adult joints and in serial sections of fetal joints.
The articular nerves are derived from the femoral, obturator,
tibial, common peroneal and recurrent peroneal nerves. Cases
recorded in the literature indicate that on rare occasions the
accessory obturator nerve also supplies the knee joint.
The femoral nerve through its saphenous branch and also
through its branches to the vastus medialis, intermedius and
lateralis muscles, supplies the suprapatellar recess, the patel-
lar periosteum, the anteromedial and anterolateral portions
of the joint capsule, the infrapatellar fat pad and vessels to
the femoral and perhaps the tibial condyles.
The tibial nerve supplies the posterior, medial and lateral
portions of the joint capsule, the infrapatellar fat pad, the
tibial periosteum, the superior tibio-fibular joint and vessels
supplying the tibial and perhaps the femoral condyles.
The common peroneal nerve supplies the anterolateral por-
tion of the capsule, the infrapatellar fat pad, the tibial peri-
osteum and vessels to lateral tibial and perhaps the lateral
femoral condyle. The recurrent peroneal nerve supplies the
tibial periosteum, the tibial tuberosity, the infrapatellar fat
pad and the superior tibio-fibular joint.
The obturator nerve supplies the popliteal vessels, the
superior part of the posteromedial capsule, the anteromedial
NERVE SUPPLY O F KNEE JOINT 129

part of the capsule, the infrapatellar fat pad and vessels to


the medial femoral condyle.
Many of the nerve fibers undoubtedly end in association
with blood vessels supplying the capsule and epiphyses. Other
types of endings could not be deduced, except for the fact that
the concentration of fibers in the posterior part of the capsule
makes it possible that proprioceptive endings are located here,
as they are in other animals studied.

ACKNOWLEDGMENT8

I wish to express my thanks to Mrs. Gail Markham for her


aid in making the microscopic sections, and to Misses Evelyn
Erickson and Geraldine Chesney for their skillful prepara-
tion of the illustrations.

LITERATURE CITED
BRUNSCHWIG, A., AND A. JUNG 1932 The importance of the periarticular in-
nervation in the pathological physiology of sprained joints. J. Bone
Joint Surg., 14: 273-276.
CRWEILHIER, J. 1844 The anatomy of the human body. The first American froin
the last Paris edition. By G. 8.Pattison. Harper and Bros., New Pork.
CUNNINGHAM’S TEXTBOOK OF ANATOMT1943 Eighth edition, edited by J. C.
Brash and E. B. Jamieson, Oxford University Press, New Pork.
DRUNER,L. 1927 Ueher die beteiligung des Nervus obturatorius an der In-
nervation des Kniegelenks. Ztschr. f. Anat. u. Entwiekl., 83: 388-390.
ECKER-WEIDERBHEIM 1896 Anatomie des Frosches. Cited by Taylor.
ELLIS, G. V. 1839-1840 Extract from a communication on the nerves of the
knee joint. The Lancet, I ; 595-597.
1882 Illustrations of dissections. From the 1876 University Col-
lege, London edition. Wood, New Pork.
FICR,R. 1904 Handbuch der Anatomie des Menschen. Edited by I(. von
Bardeleben. 8, pt. 1, sect. 1. Gustav Rscher, Jena.
GARDNER,E. 1942 Nerve terminals associated with the knee joint of the mouse.
Anat. Rec., 83: 401-419.
1944 The distribution and termination of nerves in the knee joint
of the cat. J. Comp. Neur., 80: 11-32.
1948 Conduction rates and dorsal root inflow of sensory fibers
from the knee j o i n t of the cat. Am. J. Physiol. (in press).
GHETID, V. 1939 Die Innervation der Gelenkkapseln an den Gliedmassen des
Pferdes (fur K h i k e r ) . Arch. fiir wiss. u. prakt. zierh., 75: 134-143.
GRAY’S ANATOMY OF THB HUMANRODY 1942 Twenty-fourth edition. Edited
by W. H. Lewis. Lea and Febiger, Philadelphia.
130 ERNEST GARDNER

HILTON,JOHN 1891 Rest and pain. A course of lectures. Second edition, re-
printed from the last Loiidon edition. P. W. Garfield, Cincinnati.
JAMIESON, E. B. 1903 A description of some anomalies in nerves arising from
the lumbar plexus of a foetus, and of the bilaminar musculus pectineus
found in the same foetus; with a study of the variations and relations
to nerve supply in man and some other mammals. J. Anat., 37:
~6~~286.
JELETSKY, A. G. 1931 On the innervation of the capsule and the epiphyses of
the knee joint. Vest, khir. (nos. 65-66), 82: 74-112.
LENNANDER, K. G. 1906 Ueber lokale Aniisthesie und iiber Sensibilitiit in
Organ and Gewebe, weitere Beobachtungen 11. Mitt. d. Grenz. der
Med. u. Chir., 25: 465-494.
LERICHB,RENE 1930 Recherches sur le r81e de l’innervation sensitive des
articulations. Presse Med., 38: (l), 417418.
MORRIS’HUMANANAIDMY 1942 Tenth edition, edited by J. P. Schaeffer. The
Blakiston Company, Philadelphia.
NYSTROM,GUNNAR 1917 Ueber den scherzsinn des Skelettsystems. Deut. ztschr.
f. Chir., 148: 147-168.
PATERSON, A. M. 1894 The origin and distribution of the nerves t o the lower
limb. J. Anat., 28: 84-95.
1894 The origin and distribution of the nerves t o the lower limb.
J. Amt., 88: 169-193.
PIERSOL’S HUMAN ANATOMY1930 Ninth edition, edited by G. Carl Huber.
5. B. Lippincott Company, Philadelphia.
QUAIN’S ELEMENTSOF ANATOMY1909 Eleventh edition, edited by E. A.
Schafer, J. Symington, T. H. Bryce. Longmans, Green and Co., London.
RASZEJA,F., AND J. BILLEWIOZ-STANKIEWIOZ 1934 Sur l’innervation de Is
capsule articulaire du genou chez le lapin. Abs. in Comptes Rendus de
la SOC.de Biol., 115: 1267-1268.
RUDINOER,N. 1857 Die Gelenknerven des Menschlichen K6rpers. Erlangen.
Verlag von Ferdinand Enke.
SAPPEY, PH. C. 1877 Trait6 d’anatomie descriptive. Third edition, ZZZ. Adrien
Delahaye et Cie, Paris.
TAYLOR, A. C. 1943 Development of the innervation pattern in the limb bud
of the frog. Anat. Rec., 87: 379-413.

You might also like