Muscle spindles, Golgi tendon organs, and the neural control
of skeletal muscle
JOHN N. HOWELL, Ph.D.
Ohio University
College of Osteopathic Medicine
Athens, Ohio
MARC D. BINDER, Ph.D.
University of Washington
School of Medicine
Seattle, Washington
T.
RICHARD
NICHOIS, Ph.D.
Emory University
Atlanta,
Georgia
GERALD E. LOEB, M.D.
National Institutes
of
Health
Bethe&, Maryland
The purpose ofthis report is to share with
the readershia of
JAOA.
hiahliahts of a
symposium hdd at the
30tc
~nnual
Re-
search Conference of the AOA in March
1986. The invited speakers
in
the sym-
posium were Dr.-Gerald E. Loeb of
the
National Institutes
of
Health, who also
delivered the keynote address, Dr. Man:
D. Binder of the University of Wash-
ington, and Dr.
T.
Richard Nichols of
Emory University. Dr. John N. Howell of
the Ohio University College of Os-
teopathic Medicine, program chairman
for the 1986 conference,
acted
as moder-
ator for the symposium.
In introducing the symposium, Dr.
Howell pointed out that the traditional
importance
of
somatic dysfunction and its
treatment in osteopathic practice has led
to
a natural interest within osteopathic
medicine in the aroblem of control of skel-
etal muscle fun'ction. In 1974, Korr pub-
lished an important article entitled,
"Roprioceptors and somatic dysfunc-
tion."l
In
it he related the clinical find-
ings of osteopathic practitioners to what
was
then knowh about the physiology of
the proprioceptive organs within skeletal
muscle. Specifically, he postulated that
alterations of the gain of the stretch
re
ceptor system might contribute to the
etiology of somatic dyefunction and
to
its
treatment through osteopathic manip-
ulative treatment. Korr's hypothesis has
been widely accepted within osteopathic
medicine because it provided a mecha-
nism that appeared to be consistent with
a
good deal of clinical experience.
However, as Korr pointed out in a recent
article,2 his h*thesis still awaits test-
ing in the laboratory. Since the publica-
tion of Ko~r's hypothesis, additional
information has come
to
light about the
function of muscle proprioceptors. The
goal of the symposium
was
to summarize
this newer information so that it might be
available for integration into thinking
about the pathophysiology and treatment
of somatic dysfunction.
Dr.
Binder led off wi* a summary of
the traditional ideas about how muscle
epindles and Golgi tendon organs func-
tion and then focused on newer informa-
tion about the tendon organs. On
anatomical grounds, it is clear that the
spindles are arranged parallel
to
the
ex-
trafusal muscle fibers and are thus suited
to sensing length changes of the muscle.
The tendon organs, on the other hand, are
located in series with the muscle fibers
and are thus suited to detecting force gen-
erated by the muscle fibers. Spindles have
been known for some time to be sensitive
to very small perturbetions in muscle
length, producing a reflex activation, or
shortening, of the muscle being
stretched. In contrast, tendon organs
were
thought to respond only to high lev-
els of force and to provide protection
again& the development of
too
much force
by inhibiting the homonymous muscle
(i.e., muscle of origin of the reflex).
Recent work has shown this idea about
tendon organ function to be unlikely.3.4
Each tendon organ provides a point of
insertion for 3 to
25
muscle fibers.6 Ex-
perimental stimulation of a single motor
unit, one or more fibers of which insert on
a given tendon organ, will stimulate that
organ, producing a train of spikes in
Ib
afferent neurons to the spinal cord. Stim-
ulation of a motor unit which hasno fibers
inserting on a particular tendon organ
mmetimes silences that tendon or'gan if it
has been previously activated by sus-
tained stretch of the muscle. This pre-
sumably results from unloading of the
tendon organ by virtue of contraction of
adjacent muscle fibers. Each tendon
organ appears to provide insertion for fi-
bers from motor units producing a wide
range of forces. Thus, each organ may
sample muscle activity over the full range
of force development, from the small
motor units that are the first to be re-
cruited to the large units that are re-
cruited only during maximal con-
tractions. This means that the tendon
organ input
to
the cord is not restricted
to
conditions of high force development but
can
be graded over the entire range of
muscle force.
Determination of the effect of Golgi ten-
don activity on neurons within the spinal
cord has been difficult because it has not
been possible
to
selectively stimulate the
tendon organs or their Ib afferent fibers
without at the same time stimulating the
muscle spindles or their Ia sensory fibrs.
Recently it has been shown that applica-
tion of prolonged, high-frequency, low-
amplitude vibration to muscles, which is
an effective stimulus to the muscle spin-
dles: can cause a transient elevation of
the electrical thresholds of the Ia fibers.7
Under these conditions the Ib fibers can
be
eelectively stimulated and their cen-
tral connections studied.8 Available evi-
dence indicates that activation of the Ib
sensory fibers, like the activation of cuta-
neous aiTerents, can cause either excit-
atory or inhibitory postsynaptic poten-
tials to occur in motoneurons.s The site of
integration involving the Ib input as well
as inputs from other areas within the cen-
tral nervous system appears not to be the
motoneurons, but rather interneurons in
laminae
V
and
VI.
Dr. Nichols traced the development of
theories concerning the mechanical as-
pects of the spindle-mediated stretch re-
flex and went on
to
point out that most of
our knowledge has been based on the
~tudy of single muscles. He stressed the
need for extending the analysis of stretch
reflexes to groups of synergistic muscles
acting together upon a joint. A rather ex-
treme view of the role of the stretch reflex
system was proposed in 1953 by Mertonlo
who hypothesized that normal activation
of muscle might always
be
preceded by
gamma activation of the muscle spindles,
and the resultant stimulation of Ia af-
ferent~ would in turn provide the excit-
atory input to the alpha motor neurons
and cause muscle contraction. This the-
ory was referred to as the "follow-up
length servo hypothesis." It stressed the
feedback control of muscle length and the
primacy of gamma activation for muscle
contraction. It is now known, however, for
many movements, including ballistic
types and patterned activities of a re-
petitive nature, simultaneous activation
of both gamma and alpha motor neurons
occurs, generally referred to as coactiva-
tion.11 Because the larger alpha motor
neurons conduct action potentials faster
than the smaller gamma fibers, coactiva-
tion implies that extrafusal muscle fibers
must be activated before intrafusal fibers.
Thus the initial phase of contraction
oc-
curs independently of any feedback con-
tribution from the muscle spindle system.
F'urthermore, it was pointed out that the
Golgi tendon organs might also play an
important role in feedback control, so
that not only muscle length, but some
function of both force and length would
be
controlled.l2
The latter idea
was
extended further by
Houk>3 who suggested that the param-
eter controlled bv stretch reflex might
be
stiffness, definid
as
the ratio of force
,
change
to
length change. Nichols and
Houkl4 demonstrated that the linearity
,
of the spring-like behavior of muscles in
the decerebrate cat preparation
was
lost
when the proprioceptive feedback loop
was i~terrupted, i.e. when the dorsal
continued on
page
6001117
Sept. 1986lJournal of AOAIvol. 86lno. 9
