Neuropathology and Applied Neurobiology 1986,12, 459-475

STRINGHALT I N H O R S E S : A D I S T A L AXONOPATHY

J. I . CAHILL, B. E . G O U L D E N A N D R. D. JOLLY*
Department of Veterinary Clinical Sciences and "Department of Veterinary
Pathology, Massey University, Palmerston North, New Zealand

Received for publication 30 October 1985

Revised received 2 December 1985
Accepted for publication 1 April 1986

Cahill J. I., Goulden B. E. & Jolly R. D. (1986) Neuropathology and

Applied Neurobiology 12, 459-475

Stringhalt in horses: a distal axonopathy

A detailed investigation of the neuropathology of a horse affected with

stringhalt was performed. Qualitative and quantitative light and electron
microscopy, and single teased fibre preparations of peripheral nerve
demonstrated predominantly axonal degeneration, the stage of which was
appropriate for the duration of clinical signs. There was selective
involvement of large myelinated nerve fibres. A proximal to distal increase
in the severity of pathological changes was present in the peripheral
nerves. The long left recurrent laryngeal nerve was the most severely
affected, followed in order by its right counterpart, the hindlimb and
forelimb nerves. Neurogenic atrophy of muscles innervated by affected
peripheral nerves also showed a distally graded increase in severity. No
lesions were observed in the central nervous system. It was concluded that
this disease should be classified as a distal axonopathy.

Introduction

Stringhalt, a disease of horses characterized by exaggerated flexion of the

hindlimbs, has long been recognized in epidemic and sporadic forms (Reakes,
1916; Kerrigan, 1917; Seddon & Belschner, 1926; Barry, 1956; Wheat, 1972;
Lamont, 1977; Pemberton, 1979; Pemberton & Caple, 1980; Read & Kirk, 1983,
Cahill, Goulden & Pearce, 1985). Although the aetiology of this disease is as
yet unknown, a neurotoxic pathogenesis is considered likely (Seddon &
Belschner, 1926; Barry, 1956; Lamont, 1977; Pemberton, 1979; Pemberton &
Caple, 1980; Read & Kirk, 1983; Robertson-Smith et al., 1985). The pathology
is poorly documented, but both a peripheral axonal degeneration and lesions
Address for correspondence: Dr B. E. Goulden, Department of Veterinary Clinical Sciences,
Massey University, Palmerston North, New Zealand
459
460 J. I. Cahill, B. E. Goulden & R. D . Jolly

of the spinal cord have been reported. (Lamont, 1977; Pass, 1978; Hartley,
1979; Munday, 1979; Pemberton & Caple, 1980; Robertson-Smith et al., 1985).
Traditionally, stringhalt has been considered to be a disease of the hindlimbs,
but forelimb involvement (Reakes, 1916; Cahill et al., 1985), and signs of
laryngeal dysfunction (Reakes, 1916; Kerrigan, 1917; Barry, 1956; Pemberton
& Caple, 1980) also occur.

Materials and methods

Clinical findings
To define the pathological lesions of stringhalt, a detailed investigation was undertaken on an
affected 5-year-old Thoroughbred gelding, which was 160 cm in height at the withers. This
animal was one of 21 involved in an outbreak of the disease in New Zealand (Cahill et al.,
1985).
A t the time of the initial examination, and immediately prior to killing, 6 weeks later, the
symmetry and movements of the larynx were observed through a fibreoptic endoscope.

Pathology
Muscle samples were obtained from the larynx, forelimbs and hindlimbs immediately post
mortem and frozen within 1h of collection in isopentane standing in liquid nitrogen. All
intrinsic laryngeal muscles were removed in their entirety and weighed. The forelimb muscles
sampled were the long head of the triceps brachii, extensor carpi radialis, flexor carpi ulnaris
and peroneus tertius. Muscles sampled from the hindlimbs included the vastus lateralis, biceps
femoris, gastrocnemius, extensor digitorum longus, tibialis cranialis and peroneus tertius. Two
cryostat sections (10pm) were cut from each muscle sample, one of which was stained with
haematoxylin and eosin (H&E) and the other for myosin adenosine triphosphatase (myosin
ATPase) activity at pH 9.5 using the method of Davies & Gunn (1972). The degree of pathology
present in each muscle was assessed according to the grading system adapted from Anderson
(1984), and shown in the footnote to Table 2.
Peripheral nerve samples were taken from the sites illustrated in Figures la and b. In
addition, the tibia1 nerve was sampled at the level of the junction of the gastrocnemius muscle
with its tendon. The only forelimb nerve collected was the median, at a site medial to the elbow
joint. Excluding the intrathoracic segments of the recurrent laryngeal and vagal nerves, all
nerve samples were taken from the animal anaesthetized with sodium thiopentone and glycerol
guaiacolate, and maintained on halothane and oxygen for a total of 3 h. The nerves were fixed
initially in situ in 2% glutaraldehyde for 5 min before excision. I n situ fixation was not practical
for the intrathoracic samples because of the vast volumes of fixative required. Instead,
weighted fixation was used to ensure straightening of the nerve trunk. A 200-500mg weight
was attached to the distal end of the nerve sample which was then suspended in 2%
glutaraldehyde (Dyck & Lofgren, 1966). Those to be embedded in Polarbed 812 were further
fixed for 1.5 h in glutaraldehyde, post fixed for 1.5 h in 1%osmium tetroxide, stained with 1%
paraphenylenediamine for 30 min and dehydrated through an alcohol/propylene oxide system
before embedding. Those sections destined for teased fibre preparations were further fixed for
12 min in 2% glutaraldehyde prior to immersion in the osmium tetroxide and were then stored
in glycerol.
For each nerve examined, 100 teased fibres of at least four internodes, were graded
descriptively, according to the method of Dyck (1975). Their internode lengths were measured
 Stringhalt in horses 461

Distal RLN
Vagusnerve
Midpoint RLN

Vagus nerve
-/

Proximal RLN
-3
-

vastus lateralis

-Common peroneal nerve

__-motor branch of deep
peroneal nerve
biceps femori
.- --ranch to EDL muscle
Branch of deep peroneal netv e
extensor digitoru
longus Superficial peroneal nerve

Figure 1. Sampling sites of peripheral nerves. a Recurrent laryngeal nerves (RLN); b Hind-
limb nerves. EDL. extensor digitorum longus.

using an ocular micrometer at a magnification of x 40 or x 100. Fibre diameters were

calculated from the circumference of each fibre measured by a computer digitizer on traced
enlargements ( x 750) of transverse epoxy sections of each nerve. All fibres in four fascicles
were so measured, varying from 109 in diseased nerve to 688 in nerve from control horses. The
results were plotted as histograms.
The central nervous system was removed and fixed in 12.5% formol saline. Coronal sections
of the hindbrain and selected areas of the spinal cord were embedded in paraffin, and 3 p m
sections stained with H&E, and lux01 fast blue/cresyl violet.

Results

CLINICAL FINDINGS

When examined initially, 12 weeks after the onset of symptoms, the animal
exhibited severe stringhalt action. At every step, there was marked
hyperflexion of both hindlimbs with the fetlock contacting the ventral
abdomen. Forward progression was possible only by a bunnyhopping action
with both hindlimbs. Marked atrophy of the distal hindlimb muscles was
present. Coincident with the onset of the disease, it had been noted that the
horse called less often than usual, and its whinny became quiet and hoarse.
After a further 6 weeks no clinical improvement was noted and killing was
performed.
462 J . I. Cahill, B. E. Goulden & R. D. Jolly

Table 1. Laryngeal muscle weights in a horse with stringhalt (g)

Muscle Left Right

cricoarytenoideus dorsalis 5.77 (8.17)* 5.60 (8.70)

cricoarytenoideus lateralis 3.21 (3.04) 4.09 (3.79)
cricothyroideus 5.77 (5.76) 6.56 (5.74)
arytenoideus transversus 2.75 (2.38) 2.80 (2.72)
ventricularis 1.30 (2.23) 1.68 (2.65)
vocalis 4.21 (5.10) 4.20 (5.73)

*Muscle weights of Thoroughbred horses (Quinaln et al., 1975).

At the initial examination, a visual comparison of the left and right

arytenoid cartilages revealed that left abduction during inspiration was less
than the right. Between respirations, the left cartilage was more upright and
closer to the midline. A t the subsequent examination, 6 weeks later,
inspiratory abduction of the left arytenoid cartilage was absent, whereas the
right cartilage appeared to hyperabduct. The positioning of the left arytenoid
cartilage between respirations was unchanged.

M U S C L E PATHOLOGY

At postmortem, the cricoarytenoideus dorsalis and cricoarytenoideus

lateralis muscles appeared paler than normal. Individual muscle weight
measurements (Table 1) indicated a severe decrease in the size of the
cricoarytenoideus dorsalis muscle, and a lesser decrease in the ventricularis
and vocalis muscles. Histopathological changes indicative of neurogenic
disease were observed in all intrinsic laryngeal muscles from both sides, but
were more severe in those from the left. These changes and their severity are
summarized for each muscle in Table 2, and depicted in Figure 2. The most
severe changes were noted in the cricoarytenoideus dorsalis,
cricoarytenoideus lateralis, ventricularis and vocalis muscles, and involved
extensive fascicular atrophy (Figure 2a). However, even in these severely
affected muscles, small bundles of hypertrophic fibres also existed
(Figure 2b). The least affected intrinsic laryngeal muscle was the right
arytenoideus transversus, in which the majority of fibres were normal except
at the edges of fascicles, where angular atrophic fibres were observed. No
abnormalities were noted in either cricothyroideus muscle. Myosin ATPase
staining showed fibre type grouping indicative of reinnervation, in the left
and right cricoarytenoideus lateralis and left vocalis muscles only.
Hindlimb muscles had similar histopathological changes (Figure 2c) to
those seen in the laryngeal musculature. However, pathological changes
tended to be less severe than in the larynx. In contrast, fibre type grouping
 Stringhalt i n horses 463

Table 2. Muscle pathology in a case of stringhalt

Degree of pathology

Muscle Left Right

~~

Laryngeal
cricothyroideus - -
cricoarytenoideus lateralis ++++* +++*
cricoarytenoideus dorsalis ++++ ++++
vent ricu laris ++++ +++
vocalis ++++* +++
arytenoideus transuersus +++ ++
Hindlimb
vastus lateralis - -
biceps femoris +* + +*
gastrocnemius ++ ++
peroneus tertius ++* + +*
extensor digitorurn longus + +* +*
tibialis cranialis ++ +* +++*
Forelimb
triceps brachii
extensor carpi radialis
flexor carpi ulnaris
flexor carpi radialis

The degree of pathology present in each muscle was estimated by using

a grading system adapted from that of Anderson (1984) as follows:
Mild pathology (+) Rounding of, and an increase in the number of
sarcolemmal nuclei; occurrence of a higher than normal number of
nuclei within the body of muscle fibres; minor variation in fibre size.
Moderate pathology (+ +) Fibre atrophy and hypertrophy; apparent
increase in the amount of endomysial and perimysial connective tissue;
appearance of grouped fibre atrophy.
Marked pathology ( + + +) Widespread atrophy and hypertrophy of
fibres; marked endomysial and perimysial fibrosis.
Severe pathology (+ ++
+) Widespread fatty and fibrous
replacement of muscle fibres; pyknotic nuclear clumps.
* Fibre type grouping present.

occurred more frequently, indicating increased reinnervation of previously

denervated muscle. This fibre type grouping was present in all of the distal
hindlimb muscles examined, with the exception of the gastrocnemius. In
general, pathological changes were more severe in the most distal muscles.
Forelimb muscles showed only mild lesions indicative of neurogenic disease,
but fibre type grouping was observed in the more distal muscles sampled
(Figure 2d).
464 J. I. Cahill, B. E. Goulden & R . D.Jolly

Figure 2. a The left cricoarytenoideus lateralis muscle showing fascicular atrophy, with
bundles of hypertrophic fibres also present; b The right cricoarytenoideus lateralis muscle,
showing less severe neurogenic atrophy than its left counterpart; c The left tibialis cranialis
muscle showing atrophy and hypertrophy of muscle fibres; d The right flexor carpi radialis
muscle, showing mild changes of neurogenic atrophy. Note the fibre-type grouping (a) and
atrophic muscle fibres (b) at the edges of some fascicles. Myosin ATPase, x 40.

HISTOPATHOLOGY OF PERIPHERAL NERVES

All samples of recurrent laryngeal nerves examined in transverse sections of

epoxy resin stained sections, showed a loss of myelinated fibres, with a
distally graded increase in severity (Figure 3a-c). The fibre loss was more
severe in the left nerve than in the right. Indeed, in the former nerve, only a
few myelinated fibres remained immediately caudal to the larynx (Figure 3b).
There was also widespread myelin debris present in Schwann cells and
 Stringhalt i n horses 465

Figure 3. a Left recurrent laryngeal nerve from a normal horse, illustrating the typical
density and fibre diameters; b Left recurrent laryngeal nerve immediately caudal to the larynx.
Note the marked decrease in fibre density, and the large fibre with split myelin sheath (arrow);
c Left recurrent laryngeal nerve in the area of the aortic arch. Loss of myelinated fibres and
considerable regenerative activity is evident. Many ‘onion bulbs’ and regenerating clusters are
present. d Right tibia1 nerve. There is a noticeable loss of myelinated fibres, and myelin debris
is prominent. Regenerating clusters and thinly myelinated fibres are present. Epoxy resin,
paraphenylenediamine, x 200.

macrophages, both of which were present in increased numbers. Again, this

change was progressively more severe in distal areas of the nerve. The extent
of degenerative changes at the distal end of the right recurrent laryngeal
nerve appeared similar to those of the midcervical level of the left nerve,
while the severity of changes at the right midcervical area was similar to
those at the proximal level of the left nerve.
466 J. I. Cahill, B. E. Goulden & R.D.Jolly

Table 3. Teased nerve fibre types (Dyck, 1975) in a case of stringhalt

yooffibre types*
Sample A C D E F G I Abnormal

Recurrent laryngeal nerves

Left distal 17 12 5 61 3 2 83
Left midcervical 42 7 50 1 58
Left proximal 55 9 29 2 5 45
Right distal 65 35 35
Right midcervical 57 1 42 43
Right proximal 62 4 30 (7) 2 2 38

Limb nerves
Left tibial 58 2 40 (5) 42
Right tibial 53 1 46 47
Left superficial peroneal 48 2 50 52
Left branch deep peroneal 38 61 (13) 1 62
Right branch deep peroneal 36 1 63 (11) 64
Left branch to EDL 64 36 (11) 36
Right branch to EDL 40 60 60
Right motor branch deep peroneal 42 54 (9) 2 1 57
Left common peroneal 44 54 (7) 2 56
Right common peroneal 46 54 54
Left branch to BF 54 2 44 46
Right branch to BF 66 2 2 30 34
Right median nerve 48 5 38 4 2 3 52

EDL, extensor digitorum longus muscle.

BF, biceps fernoris muscle.
*A, normal appearance; C, segmental demyelination; D, segmental demyelination and
remyelination; E, axonal degeneration; F, segmental remyelination; G, focal thickenings of
myelin; I, fibre with several proximal internodes with or without proximal paranodal or
segmental demyelination, and distal to those degeneration (Dyck, 1975).
Numbers in brackets, E fibres with large myelin ovoids.

Regenerating clusters of nerve fibres, and fibres with disproportionately

thin myelin sheaths in relation to their axonal calibre, were more common in
the proximal parts of affected nerves (Figure 3c). ‘Onion bulb’ formations,
usually surrounding thinly myelinated fibres, were also more frequently
observed in proximal areas.
The hindlimb nerves exhibited similar pathological changes to those
observed in the recurrent laryngeals (Figure 3d). The same distally
progressive increase in severity was present, although overall the pathology
was less severe. While not present in every section, Renaut bodies were a
common finding. The median nerve of the forelimb also showed similar
changes, but loss of myelinated fibres was not apparent by subjective
assessment.
 Stringhalt in horses 467

Figure 4. a A teased fibre from the left recurrent laryngeal nerve in the midcervical area.
Myelin debris is associated with remnants of myelinated fibres, indicating axonal degeneration
of some weeks duration; b A teased fibre from the right deep peroneal nerve. Recent axonal
degeneration is evidenced by the linear row of large myelin ovoids x 100.

TEASED FIBRE PREPARATIONS OF PERIPHERAL NERVES

Teased myelinated fibres, graded according to the scheme of Dyck (1975),

showed a high incidence of abnormalities in all nerves examined (Table 3).
This ranged from 34% abnormal fibres in the branch of the peroneal nerve
innervating the biceps fernoris, to 83% in the left recurrent laryngeal nerve
immediately caudal to the larynx. The predominant abnormality observed
was that of axonal degeneration (type E), of which two variations were
observed. Small fragments of myelin debris were scattered along the vestiges
of each myelinated fibre (Figure 4a). Other fibres appeared as continuous
rows of large myelin ovoids, (Figure 4b).
Changes indicative of demyelination (type C) and remyelination (type F),
and combinations of these two processes (type D) were observed, but less
frequently than type E changes. The demyelination was usually paranodal in
distribution, while remyelinated segments appeared as short intercalated
internodes. Only three nerves contained fibres with type I change, in which
there is paranodal demyelination or normal nodal architecture proximal to
myelin ovoids. Paranodal demyelination was usually observed in these fibres.

ELECTRON MICROSCOPYOF PERIPHERAL NERVES

Ultrastructural examination of nerves extended the light microscopic

findings. Denervated Schwann cells, (Figure 5a) were observed in all nerves
with the exception of the median and the branch of the peroneal innervating
the biceps fernoris. Myelin debris was seen within Schwann cell cytoplasm,
Schwann cell processes and in macrophages. This ranged from large
lamellated my elin ovoids to complex osmiophilic structures of varying
468 J. I . Cahill, B. E. Goulden & R. D.Jolly

Figure 5. Electron micrographs from recurrent laryngeal and hindlimb nerves. a A band of
Bungner containing myelin debris in several stages of degradation x3400; b A band of
Bungner containing a multiloculated structure (arrow), thought to be a result of myelin
degradation x 5200; c A fibre from a regenerating cluster, with accumulation of membranous
dense bodies in its axoplasm x 15300. d A thinly myelinated fibre with surrounding Schwann
cell processes. Note the excess convoluted basement membrane and the protagon granule
in the outer Schwann cell cytoplasm x7800. e A cluster of thinly myelinated regenerating
fibres. Note the apparently atrophic fibre (arrow), with small axonal calibre relative to its
myelin sheath thickness x 5200.

morphology, often with a multiloculated appearance (Figure 5a and b). In

addition smaller elongate structures, known as protagon granules
(Figure 5d), and lipid droplets or myelin bodies were commonly seen in
Schwann cell cytoplasm. A small proportion of fibres which had
 Stringhalt in horses 469

disproportionately thick my elin sheaths relative to their axonal calibre, were

considered to be atrophic (Figure 5 e).
Evidence of regeneration was provided by the presence of nonmyelinated
axonal sprouts prior to myelination and clusters of regenerating fibres
(Figure 5e). Excess convoluted basement membrane was seen in association
with the latter, and with thinly myelinated fibres (Figure 5d). Less commonly
seen were early ‘onion bulb’ formations. Regenerating axonal sprouts in some
instances had a lucent floccular appearance, while others contained densely
packed disoriented neurofilaments. Other abnormalities noted were the
presence of dense membranous bodies (Figure 5c), the margination of
microtubules, and distension of the adaxonal Schwann cell cytoplasm.

MORPHOMETRY O F P E R I P H E R A L N E R V E S

The distribution of fibre diameter measurements of myelinated fibres in the

distal recurrent laryngeal nerves from the horse with stringhalt and from six
control horses (Cahill, 1985) is presented in Figure 6. In the nerve from the
horse affected with stringhalt there has been a shift to smaller diameters. The
large fibre peak was reduced in height while the small peak was increased,
giving the distribution a more bimodal appearance than is normal for
recurrent laryngeal nerve. Similar findings were observed in the midcervical
and proximal recurrent laryngeal nerves from the animal with stringhalt.
This selective loss of large diameter fibres was reflected by a decrease in
mean fibre diameters.
Internode lengths of nerve fibres from the recurrent laryngeal nerves of
the horse with stringhalt are presented in Table 4. In general, there was little
difference between control values and those from the affected animal.

4
: 8
I
in
u<l
12
Fibre diameter ( p m )
Figure 6. Histograms of the myelinated nerve fibre diameter distributions of the recurrent
laryngeal nerve immediately caudal to the larynx. a control horses, left side (Cahill, 1985),
mean fibre diameter ( a ) 8-4;b horse with stringhalt, left side = 6.4; c horse with stringhalt,
right side, = 7.5.
470 J. I. Cahill, B . E. Goulden & R. D. Jolly

Table 4. Mean internode length of nerve fibres of recurrent

laryngeal nerve from a horse with stringhalt

Mean internode Coefficient of

length variation (%)
Nerve level 04
Left distal 732 (865)* 10.0 (8.3)
Left proximal 1069 (975) 7.3 (7.1)

Right distal 755 (828) 9.8 (7.8)

Right proximal 935 (975) 8.3 (7.1)

( )* Values for control horses (Cahill, 1985)

Presumably this reflected the low incidence of intercalated internodes in the

nerve fibres from the horse with stringhalt. In addition, the fibres being
measured were those which were unaffected by the disease process.

HISTOPATHOLOGY OF THE CENTRAL NERVOUS SYSTEM

The brain and spinal cord were essentially normal, except for a small number
of axonal swellings (spheroids) a t various levels between the medulla
oblongata and cauda equina.

Discussion

The pathological investigation of this horse with stringhalt, showed the

existence of a distal axonopathy, greatest in degree in the nerve known to be
the longest in the horse (Duncan & Griffiths, 1973), the left recurrent
laryngeal, and least in the shortest nerves studied. The predominant finding
in all histological preparations was one of axonal degeneration, indicating a
primary effect of the disease on the axon. Moreover, in the affected nerves, a
proximal to distal increase in the severity of damage was noted. Quantitative
studies confirmed this distally distributed, selective loss of large diameter
myelinated fibres. No conclusive evidence of involvement of long nerve fibres
within the central nervous system was found.
Histological examination of muscles supplied by these peripheral nerves,
demonstrated neurogenic atrophy of varying degrees, with some regeneration
as shown by fibre type grouping. The degree of muscle pathology followed the
same pattern observed in the nerve sections, and was directly related to the
length of nerve supplying the particular muscle.
The presence of pathological changes in the left recurrent laryngeal
nerves of horses is extremely common, and is related to the disease equine
laryngeal hemiplegia (Duncan, 1975; Anderson, 1984; Cahill, 1985).
 Stringhalt in horses 471

Consequently, the possibility that alterations were present in the left nerve
before the onset of stringhalt should be considered. However, in this case of
stringhalt the pattern of pathological changes in the laryngeal muscles and
nerves differed from those described in equine laryngeal hemiplegia (Duncan,
1975; Anderson, 1984; Cahill, 1985).
The pathological findings in the peripheral nerves were in keeping with the
4 month duration of the disease. Of the type E degenerated fibres observed in
teased fibre preparations, the vast majority consisted of filamentous strands
of tissue with small remnants of myelin debris spaced out along them. Thus, it
seemed that consequent to the initial axonal degeneration, the process of
myelin degradation and the removal of debris was almost complete, a process
known to take as long as several months (Nathaniel & Pease, 1963). Light and
electron microscopic findings also confirmed the predominance of these
chronic changes. However, in all histological preparations, a low incidence of
active axonal degeneration was also present. This was indicated by the
following observations: (i) Type E teased fibres were apparent as large myelin
ovoids. These fibres indicate an early stage of fragmentation of the myelin
sheath and axis cylinder in the process of axonal degeneration (Dyck et al.,
1968). (ii) Ultrastructurally, Schwann cells were observed containing large
myelin ovoids, in which the lamellar pattern of myelin was preserved. This
pattern is supposedly retained for approximately 6 days after degeneration
commences (Nathaniel & Pease, 1963). (iii) Apparently degenerating axonal
sprouts, were seen on electron microscopic examination, (iv) Type I teased
fibres were present, with early distal degeneration, and usually proximal
paranodal demyelination on the same fibre. These fibres have been reported
in a number of distal axonopathies, including uraemic neuropathy (Dyck
et al., 1971), thiamine deficiency (Takahashi & Nakamura, 1976) and
acrylamide intoxication (Hopkins, 1970). It has been suggested that they have
been sampled at the site in the fibre, distal to the level where axonal
degeneration has occurred. Proximal to this point paranodal demyelination
may be seen, secondary to the primary pathological changes within the axon
(Dyck et al., 1971).
A number of ultrastructural alterations were observed, which supported
the primary axonal nature of the disease. These were as follows: (i)
Degeneration of axonal sprouts, unmyelinated and myelinated fibres was
present. (ii) Accumulation and rearrangement of axoplasmic organelles was
observed. These have been reported in a number of distal axonopathies,
including hexacarbon intoxication, giant axonal neuropathy and buckthorn
intoxication (Asbury et al., 1972; Spencer et al., 1980; Heath et al., 1982). (iii)
There were projections of adaxonal Schwann cell cytoplasm extending into
the axoplasm and a distension of the adaxonal Schwann cell cytoplasm.
These may indicate attempts by the Schwann cell to sequester excess or
abnormal organelles from the axon (Spencer & Thomas, 1974; Spencer &
Schaumburg, 1976). (iv) Large numbers of collagen pockets were observed.
472 J. I. Cahill, B. E. Goulden & R. D.Jolly

These are seen in many chronic neuropathies, and are thought to reflect the
loss of unmyelinated fibres (Thomas, 1973). (v) There were increased numbers
of myelin bodies and protagon granules in Schwann cell cytoplasm, features
of axonal degeneration and regeneration (Schroder, 1975; Landon & Hall,
1976).
The ability of the peripheral nerve to regenerate, even in the presence of
continued degeneration, was reflected by the frequent appearance of
regenerating clusters. However, there was some evidence to indicate that not
all the attempts at regeneration were successful. Continued degeneration and
demyelination were apparent in some fibres of the clusters. Some of these
fibres contained lamellated myelin bodies in their Schwann cell cytoplasm,
and in others accumulated organelles, indicative of local deranged axonal
transport, were seen. The presence of atrophic nerve fibres in regenerated
nerve is thought to represent the regression of regenerated fibres, following
their failure to reinnervate the end organ (Schroder, 1972). The atrophic
fibres observed in this investigation probably resulted from the above
process.
Evidence of segmental demyelination and remyelination was provided by
the existence of paranodal demyelination or intercalated internodes, which
was most obvious in the recurrent laryngeal nerves. Presumably these
changes were consequent to axonal pathology, representing secondary
segmental demyelination as seen in the type I fibres. Structures generally
associated with repetitive episodes of segmental demyelination and
remyelination of nerve fibres, ‘onion bulbs’ (Webster et al., 1967; Dyck, 1975;
Schroder, 1975; Madrid, Bradley & Davies, 1977), were also noted in a number
of the nerves examined.
The abundance of fibres with disproportionately thin myelin sheaths in a
number of the recurrent laryngeal and limb nerve samples could indicate
either regenerated or remyelinated nerve fibres. It seems likely that they
represented the former, because of the high incidence of degeneration and the
low incidence of demyelination noted in teased fibre preparations.
Recovery from stringhalt will depend on the ability of axons to
regenerate. Because of the long distances over which the regenerating axons
must grow, recovery will be a protracted and perhaps incomplete process. For
instance, in this case, regenerating clusters were observed at the aortic arch.
From here they would have to grow along the length of the recurrent
laryngeal nerve to reinnervate the appropriate intrinsic laryngeal muscle, a
distance of some 100cm in the adult horse (Duncan & Griffiths, 1973).
Moreover, while regeneration of nerve fibres, and thus recovery of function is
possible, there may be lasting changes in the muscles. Alterations in myosin
ATPase fibre types do occur, and could permanently affect muscle function.
In a competitive animal such as the racing Thoroughbred, this could result in
impairment of performance.
Although the pathological changes were most severe in the long recurrent
 Stringhalt in horses 473

laryngeal nerves, the predominant clinical sign of stringhalt is hyperflexion

of the hindlegs. An explanation for this apparent discrepancy between
pathological and clinical findings may be related to differing functions of the
largest nerve fibres of the hindlimb and of the laryngeal nerves. In the former
the large, vulnerable nerve fibres come from the muscle spindles (Bowsher,
1962). Thus, damage to these could produce the characteristic exaggerated
flexion of the hindlimbs seen early in the course of the disease. Whereas in
the recurrent laryngeal nerves the largest fibres innervate the left
cricoarytenoideus lateralis muscle (Cahill, 1985), and damage to these nerve
fibres may not be manifested clinically.
When considering the aetiology of stringhalt, many causative factors are
known to produce distal axonopathies in other species of animals and man.
These include toxins, vitamin deficiencies and metabolic dysfunction
(Cavanagh, 1979). The possibility of a toxin being the main aetiological agent
in this disease cannot be ignored. Stringhalt most often occurs in certain
climatic conditions which are conducive to mycotoxin production on soil and
pasture (Pemberton & Caple, 1980). Moreover, this disease is frequently seen
in association with the plant Hypochaeris radicata, which may contain a
neurotoxic agent. In addition, since the disease occurs under conditions in
which a very low plane of nutrition is available to the grazing animal, a
deficiency, such as that of B vitamins, may play a part in its pathogenesis. It
is also possible that stringhalt may result from a combination of the above
factors, or that some other factors may be implicated. It may be possible that
more than one pathological process may result in the clinical signs of
stringhalt. For this reason, and to delineate the full range of pathological
changes which may occur with this disease, further investigations of cases of
stringhalt should be performed.

Acknowledgements

The authors are grateful to the New Zealand Equine Research Foundation
for financial support, to Mrs P. Davey for technical assistance, Professor R.
Munford for help with statistics and computing, Mr T. Law for the
production of the photographs and Mr R. Bennett for production of the
electron micrographs.

References

ANDERSON L.J. (1984) A study of some muscles of t h e equine larynx and soft palate Ph.D.
Thesis, Massey University
ASBURYA.K., GALE M.K., COX S.C., BARINCERJ.R. & BERG. B.O. (1972) Giant axonal
neuropathy -a unique case with segmental neurofilamentous masses, Acta
Neuropathologica 20, 237-247
474 J. I. Cahill, B . E. Goulden & R. D. Jolly

BARRYW.C. (1956) The incidence of Australian stringhalt in horses in New Zealand. New
Zealand Veterinary Journal 4, 26-27
BOWSHER D. (1962) Zntroduction to Neuroanatomy. Blackwell Scientific Publications, Oxford
CAHILLJ.I. (1985) The pathology of laryngeal hemiplegia in the horse. Ph.D. Thesis, Massey
University
CAHILLJ.I., GOULDENB.E. & PEARCE H.G. (1985) Stringhalt. A review and some observations.
New Zealand Veterinary Journal 33, 101-104
CAVANAGH J.B. (1979) The dying back process. A common denominator in many naturally
occurring and toxic neuropathies. Archives of Pathology & Laboratory Medicine 103,
659-664
DAVIESA S . & GUNNH.M. (1972) Histochemical fibre types in the mammalian diaphragm.
Journal of Anatomy 112, 41-60
DUNCAN I.D. (1 975) The pathology of equine laryngeal hemiplegia. Ph.D. Thesis, University of
Glasgow
DUNCANI.D. & GRIFFITHSI.R. (1973). Pathological changes in equine laryngeal muscles and
nerves. Proceedings of the 19th Annual Convention of the American Association of Equine
Practitioners. 97-113
DYCK P.J. (1975) Pathological alterations of the peripheral nervous system of man. In
Peripheral Neuropathy. Eds P.J. Dyck, P.K. Thomas & E.H. Lambert. W.B. Saunders & Co,
Philadelphia. 296-336
DYCKP.J., GUTRECHT J.A., BASTRON J.A., KARNESW.E. & DALEA.J.D. (1968) Histological and
teased fibre measurement of sural nerve in disorders of lower motor and primary sensory
neurones. Mayo Clinic Proceedings 43,81-123
DYCKP.J., JOHNSON W.J., LAMBERTE.H. & O’BRIEN P.C. (1971) Segmental demyelination
secondary to axonal degeneration in uremic neuropathy. Mayo Clinic Proceedings 46,
400-431
DYCKP.J. & LOFGRENE.P. (1966) Method of fascicular biopsy of human peripheral nerve for
electrophysiologic and histologic study. Mayo Clinic Proceedings 41, 778-784
HARTLEY W.J. (1979) Personal communication. Cited by Pemberton D.H. & Caple I.W. (1980).
Australian stringhalt i n horses. Veterinary Annual 20, 167-171
HEATHJ.W., UEDA S., BORNSTEINM.B., DAVIES G.D. & RAINE C.S. (1982) Buckthorn
neuropathy i n vitro: evidence for a primary neuronal effect. Journal of Neuropathology and
Experimental Neurology 41, 204-220
HOPKINSA. (1970) The effect of acrylamide in the peripheral nervous system of the baboon.
Journal of Neurology, Neurosurgery and Psychiatry 33, 805-816
KERRIGANJ. (1917) So-called Australian stringhalt of horses. New Zealand Journal of
Agriculture 15, 20-23
LAMONTB. (1977) ‘Dandelion’ stringhalt in horses. University of Sydney Post-Graduate
Foundation i n Veterinary Science, Control and Therapy, 475.
LANDON D.N., & HALLS. (1976) The myelinated nerve fibre. In The Peripheral Nerue. ed D.N.
Landon, Chapman & Hall, London. 1-105
MADRIDR., BRADLEY W.G. & DAVIESC.J.F. (1977) The peroneal muscular atrophy syndrome.
Clinical, genetic, electrophysiological and nerve biopsy studies. Part 2. Observations on
pathological changes in sural nerve biopsies. Journal of the Neurological Sciences 32,
91-122
MUNDAY B.L. (1979) Personal communication. Cited by Pemberton D.H. & Caple I.W. (1980)
Australian stringhalt in horses. Veterinary Annual 20, 167-171
NATHANIEL E.J.H. & Pease D.C. (1963) Collagen and basement membrane formation by
Schwann cells during nerve regeneration. Journal of Ultrastructural Research 9, 550-560
PASS D.A. (1978) Personal communication. Cited by Pemberton D.H. & Caple I.W. (1980).
Australian stringhalt in horses. Veterinary Annual 20, 167-171
PEMBERTON D.H. (1979) Investigations of a recent outbreak of stringhalt in horses in East
Gippsland. Victorian Veterinary Proceedings 37, 16
 Stringhalt in horses 475

PEMBERTON
D.H. & CAPLEI.W. (1980) Australian stringhalt in horses. Veterinary Annual 20,
167-171
QUINLANT.J., GOULDENB.E., & DAVIESA.S. (1975) Bilateral asymmetry of equine laryngeal
muscles. New Zealand Veterinary Journal 23, 145-147
READ D.H. & KIRKJ. (1983) Outbreak of stringhalt. Surveillance Animal Health Division of The
New Zealand Ministry of Agriculture and Fisheries 10, 22-23
REAKESC.J. (1916) Notes on a form of nerve derangement in horses. New Zealand Journal of
Agriculture 12, 364-368
RENAUT J. (1881) Recherches sur quelques points particuliers de l’histologie des nerfs. 1. La
gaine lamelleuse et la systeme hyalin intravaginal. Archives de Physiologie Normale et
Puthologique 8, 161-190
ROBERTSON-SMITH R.G., JEFFCOTT, L.B., FRIENDS.C.E. & BADCOEL.M. (1985) An unusual
incidence of neurological disease affecting horses during a drought. Australian Veterinary
Journal 62, 6-12
SCHRODER J.M. (1972) Altered ratio between axon diameter and myelin sheath thickness in
regenerated nerve fibres. Brain Research 45, 49-65
SCHRODER J.M. (1975) Degeneration and regeneration of myelinated nerve fibres in
experimental neuropathies. In Peripheral Neuropathy. Eds P.J. Dyck, P.K. Thomas & E.H.
Lambert. W.B. Saunders & Co., Philadelphia. 337-362
SEDDON H.R. & BELSCHNER H.G. (1926) Stringhalt in horses. Agricultural Gazette of New South
Wales 31, 381-382
SPENCER P.S. & SCHAUMBURG H.H. (1976) Central-peripheral distal axonopathy-the pathology
of dying-back polyneuropathies. In Progress in Neuropathology. Ed H.E. Zimmerman.
Grune and Stratton, New York. Vol. 3, 253-295
SPENCER P.S. & SCHAUMBURG H.H. (1977) Ultrastructural studies of the dying-back process. 3.
The evolution of experimental peripheral giant axonal degeneration. Journal of
Neuropathology and Experimental Neurology 36, 276-299
SPENCER P.S., SCHAUMBURG H.H., SABRIM.I. & VERONESI B. (1980) The enlarging view of
hexacarbon neurotoxicity . CRC CriticaE Reviews in Toxicology 1, 278-356
SPENCER P.S. & THOMAS P.K. (1974) Ultrastructural studies of the dying-back process. 2. The
sequestration and removal by Schwann cells and oligodendroglia of organelles from
normal and diseased axons. Journal of Neurocytology 3, 763-783
TAKAHASHI K. & NAKAMURA H. (1976) Axonal degeneration in beri-beri neuropathy. Archives of
Neurology 24, 358-364
THOMAS P.K. (1973) The ultrastructural pathology of unmyelinated nerve fibres. In New
Developments in Electromyography and Clinical Neurophysiology Karger, Basel Vol. 2,
227-239
H.DEF.,SCHRODER
WEBSTER A.K. & ADAMSR.D. (1967) The role of Schwann cells
J.M., ASBURY
in the formation of ‘onion bulbs’ found in chronic neuropathies. Journal of Neuropathology
and Experimental Neurology 26, 276-299
WHEAT J.D. (1972) In Equine Medicine and Surgery. Eds E.J. Catcott & J.F. Smithcors. 2nd ed.
American Veterinary Publications, Illinois 567-568
I