F • CLINICAL ASPECTS OF SPINAL CORD INJURIES
235
conduction of the spinal cord has been completely destroyed or not. This has led to the
conception that paraplegia-in-flexion is pathognomonic of complete transection of the
cord while paraplegia-in-extension is pathognomonic of incomplete lesions. Moreover,
Fulton, Liddell & Rioch's animal experiments (19303, b) indicated that the extensor
hypertonus depends on the transmission of postural reflexes along intact vestibulo-spinal
and probably also ventro-reticulo-spinal pathways to the cord centres below the level
of the lesion and only destruction of these tracts results in a preponderance of the flexor
reflexes and paraplegia-in-flexion. However, experimental and clinical experiences have
shown that Riddoch's theory is no longer tenable.
Sherrington found a variability in the order of recovery of the various spinal reflexes
from spinal shock in the dog, following complete transection of the cord, and in some
individuals a preponderance of extensor rigidity was evident: cThe limbs are kept
extended at knee and ankle to a degree that it is difficult to break through by the inhibition
accompanying elicitation of the flexion reflex stimulation of the foot'. Sherrington's
explanation was that 'some incidental circumstances determining the preponderance
of some passive attitude of the limb during the early days succeeding the lesion may, by
its influence on the interaction of the recovering spinal arcs, impress an unwanted reflex
habit about the limb 5 . Denny Brown & Liddell (1927, 1928) found that the predominant
flexor reflex-activity in the stage of reflex automatism of the isolated cord following tran
section in the cat and dog was replaced in time by such strong extensor spasticity and
rigidity that the animals were able to stand. In this connection, it may be remembered that
Riddoch (1917) himself had found that, in later stages of heightened reflex activity of
complete lesions, the difference in the tone of the extensor and flexor groups (the latter
being predominant) becomes less marked. Foerster (1936) reported two cases of tran
section of the cord of over one year's duration, with preponderance of extension reflex
synergy. In one of these cases, the extensor reflex of the legs could be elicited in both
legs by stimuli applied to the ano-genital region. Further proof of the great variability
in the reflex activity was given by Elkins & Wagner (1946), Scarff & Poole (1946), ScarfF
(1952), Dick (1949) and Kuhn (1950). The latter author found 19 amongst 28 patients
with complete transection at levels between T2 and Ti2 who had extensor spasms
predominant over flexor spasms.
The great number of traumatic paraplegic survivors from the First World War
prompted me to study the important problem of factors influencing the patterns of
reflex function in the spinal man (Guttmann, 1946, 1952, 1953, I 954> I 97°)- There is
no doubt that septic conditions due to pressure sores and urinary infection as well as
contractures of muscles and joints maintain and increase the predominant flexor synergy.
Moreover, there can be no doubt about the great influence which the positioning of the
paralysed limbs during the early stages of paraplegia has on the development of the
patterns of reflex synergies in later stages, in both complete and incomplete lesions of the
spinal cord. In particular, prolonged fixation of the paralysed limbs in adduction and
semi-flexion, as is so often produced by the habit of nurses placing a pillow or other
support under the knees, invariably promotes predominant flexor synergy and conse
quently paraplegia-in-flexion (Guttmann, 1946). The explanation is that the constant
approximation of the insertion points of the flexors of hips and knees in such cases
