larly long period of survival of the virus in the mucous membrane
of affected human beings has, however, not been demonstrated.
Epidemics rise and fall, and rarely disappear suddenly. The
series of cases that form the basis of the present study constitute
the instances still arising that owe their initial impulse to the epi-
demic wave that appeared in New York in 1909, and has since been
slowly disappearing.
"Neustaedter and Thro, N. F. Med, Jour,, 191 1, xciv, 813.
* Osgood and Lucas, Jour. Am. Med. Assn., 191 1, Ivii, 495.
Digitized by
Google
14 A Clinical Study of Acute Poliomyelitis.
NEUTRALIZATION TEST.
As has been emphasized in the discussion of the epidemiology of
poliomyelitis, abortive attacks are of common occurrence when the
disease is epidemic. These instances of poliomyelitis are difficult
to recognize clinically and a specific biological test would be of
great assistance in the diagnosis of such atypical infections. Up
to the present time no characteristic reaction has been demonstrated
during the acute stage of the disease. Levaditi and Landsteiner,
Romer and Joseph, and Flexner and Lewis have shown that the
serum of recovered monkeys is able to protect normal animals
against fatal doses of the virus. Netter and Levaditi, Flexner and
Lewis, and Anderson and Frost have shown that the serum of
recovered human cases possesses the same protective property. The
last named observers demonstrated this quality in the sera of 66.7
per cent, of a series of abortive instances of poliomyelitis. They
also proved that normal human serum has some neutralizing action,
but are of the opinion that this action has quantitative limits
which clearly differentiate it from the action of the serum of persons
who have had poliomyelitis. While such protective bodies do not
appear in the serum until about two weeks after the onset of the
disease, the fact that the virus is known to survive in the throats of
•experimental animals for a much longer period of time may make
it advisable in extensive epidemics to employ the test in order that
the community may be protected from suspected individuals who
may act as carriers.
Up to the time of the publication of the paper by Anderson and
Frost, the reliability of the test was not questioned. These ob-
servers show that normal human serum possesses limited power to
neutralize the virus of poliomyelitis. During. the past year at the
Hospital of The Rockefeller Institute, the serum of a number of
individuals was tested for its ability to protect monkeys against in-
fection with the virus of poliomyelitis. The material consisted
of sera from normal persons, from those exposed to infection in the
hospital ward, from individuals suspected of having an abortive
attack of the disease, and from a certain number of typical cases.
The test was performed in the following way : the serum was mixed
with a fatal dose of a known active virus, incubated from one
Digitized by
Google
Francis W. Pedbody, George Draper, and A. B. Dochez. 15
to two hours at 37° C. and then allowed to stand for twenty-four
hours on ice. The injections were made intracerebrally into
monkeys (Macacus rhesus). The protocols of the experiments
follow.
NORMAL SERA.
Experiment /.—The serum was obtained from a normal adult, one of the
physicians in constant attendance on cases of poliomyelitis in the hospital ward.
During this time he had presented no signs of illness. One c.c. of the fresh
serum was mixed with o.i c.c. of virus and was incubated for two hours at 37** C.
After standing over night on ice, the mixture was injected intracerebrally into
a monkey. After seventeen days the animal began to show signs of weakness
and was found dead on the following day. At autopsy the characteristic lesions of
poliomyelitis were present in the spinal cord.
Experiment II. — In this instance the serum was obtained from a normal adult,
a nurse in charge of cases of poliomyelitis in the hospital. She had no paralyses
and gave no history of any illness resembling poliomyelitis, nor was she ill at
any time during her service in the poliomyelitis ward. One c.c. of this serum
was mixed with o.i c.c. of virus and neutralization tested in the manner
described. The monkey inoculated became ill on the eighth day after injection.
It was somewhat depressed and its movements were slow for several days. On
the nineteenth day it developed diarrhea and eight days later it was found dead.
At no time did the animal show definite paralyses. At autopsy the microscopic
lesions were not characteristic.
Experiment III, — ^The serum for this experiment was obtained from a girl of
eight years of age, a patient in the hospital suffering from chronic endocarditis.
She had no paralyses and gave no history of any illness resembling poliomyelitis.
Two tests were made with this serum. The first monkey was inoculated with a
mixture of i cc. of fresh serum and 0.1 c.c. of virus, prepared in the usual way.
The animal survived and at no time developed any symptoms. There was no
hyperexcitability and no paralysis. A second test was made with this serum, in
which I cc. of serum was mixed with 0.3 c.c. of virus. This animal likewise
remained healthy and survived.
Experiment IV, — ^The serum for this experiment was obtained from one of
the doctors in constant attendance on the ward. He gave no previous history of
poliomyelitis, but some six weeks previously had a slight illness of about a
week's duration, the chief symptoms of which were diarrhea, dizziness, and pain
in the neck. There was no muscular weakness. One cc. of this serum was
mixed with 0.1 c.c. of virus and the neutralizing power tested in the usual
manner. Seven days after injection, the animal's nutrition was poor; there were
no paralyses and no hyperexcitability. On the fortieth day the monkey was still
living, when it was again inoculated with i c.c. of serum mixed with 0.3 c.c of
virus. The animal developed diarrhea, became emaciated, and died on the
eighteenth day after the second injection, without having developed any
paralyses.
Experiment F.— The serum of a normal adult, a nurse in attendance on cases
of poliomyelitis, was used for this experiment. Some weeks previously she had
Digitized by
Google
16 A Clinical Stvdy of Acute Poliomyelitis.
a slight illness, lasting three days, the symptoms of which were dizziness and
gastro-intestinal disturbance. There was no muscular weakness and no pain
associated with the attack. One c.c. of this serum was mixed with o.i c.c. of
virus and after allowing time for neutralization to occur was injected intra-
cerebrally into a monkey. Forty days later the animal was living and had shown
no symptoms. At this time another test was made, using i c.c. of serum and
0.3 cc of virus. Eleven days later the animal's movements became slow, it
grew thin and weak and died on the fourteenth day after the second injection.
At autopsy the grey matter of the cord looked somewhat edematous. There
were no definite microscopic lesions of poliomyelitis.
Experiment VI. — ^The serum for this experiment was obtained from M. S.,
two and one half years old, a patient in the hospital suffering from cerebral
hemorrhage. Two tests were made with this serum ; in one, i cc. of serum was
mixed with o.i c.c. of virus; and in the other, i c.c. of serum with 0.3 cc. of
virus. The first animal died of poliomyelitis in seven days and the second in
eight days.
SERA OF SUSPECTED CASES.
Experiment VII, — The serum was obtained from Patrick T., aged twenty-
three months, a patient in the poliomyelitis ward. At the time of admission to
the ward, the patient's sister was suffering from poliomyelitis with paralyses.
The patient on admission was drowsy, showed the characteristic irritability on
being disturbed, and had slight fever. The spinal fluid showed 94 cells per cmm.,
practically all of which were of the mononuclear type. Although the mother
noticed some weakness in the child when he got up from a sitting posture, the
child showed no paralyses while in the hospital. One cc of the serum obtained
from this patient seven days after the onset of the disease, was mixed with 0.1
cc. of active virus and injected intracerebrally into a monkey. Nine days later
the animal died of typical poliomyelitis.
Experiment VIIL — ^Thc serum for this test was obtained from Donald P., the
brother of a patient who died of poliomyelitis in the hospital ward. He had
shown suspicious symptoms at the time of the other child's illness. The serum
was obtained about three weeks after the suspected attack and i c.c. of it was
mixed with 0.1 cc. of virus and injected intracerebrally into a monkey. Ten
days later the animal died, having developed a characteristic poliomyelitis.
Experiment IX, — ^The serum used in this test was obtained from H. B., a
patient in the hospital A constant playmate had died of poliomyelitis some two
wTeeks before the patient's admission. The patient, when studied in the ward, had
slight fever and showed marked drowsiness. There were no paralyses but slight
spasticity on walking and some rigidity of the neck. The maximum cell count
in the spinal fluid was 62 cells per cmm., chiefly mononuclear. The globulin
content of the fluid was slightly increased. A monkey, injected intracerebrally
with a mixture of 0.1 cc. of active virus and i c.c. of the patient's serum, ob-
tained three weeks after the onset of the illness, showed no signs of poliomyelitis.
SERA OF TYPICAL CASES OF POUOMYELITIS.
Experiment X, — For this experiment, serum was obtained from C T., an
aduh who had suffered from poliomyelitis thirty years previously. Two tests
Digitized by
Google
Francis W. Peahody, George Draper, and A. B. Dochez. 17
were made with this serum. In one, o.i cc of active virus was mixed with i c.c.
of scrum; and in the other, 0.3 cc. of the virus was mixed with i cc of serum.
Two monkeys were injected with the material in the usual way. Both animals
were protected, neither showing any signs of poliomyelitis.
Experiment X/.— In this insUnce the senim was obtained from an individual,
E. R., sixteen years of age, who had had poliomyelitis fourteen years pre-
viously. One c.c of serum was mixed with o.i c.c of active virus and injected
into a monkey. Twenty days later the animal developed a paralysis of one leg.
The paralysis, however, did not progress and the condition of the monkey
returned to normal, with the exception of a residual paralysis. Inasmuch as the
dose of virus given has invariably proven fatal in unprotected monkeys, the
serum of this patient must be considered as having some protective power*
Experiment XIL — The serum for this test was obtained from C. A., a child
eleven years old who had had poliomyelitis ten years previously. One cc of
serum was mixed with 0.1 cc of active virus and injected into a monkey in the
usual way. The animal was protected, showing no symptoms of poliomyelitis at
any time.
Experiment XIII. — ^The serum used for this experiment was obtained from
M. G., four years old, a patient in the poliomyelitis ward. On admission there
was partial paralysis of the external ocular muscles, high fever, and some stiff-
ness of the neck. The maximum cell count in the spinal fluid was 320 cells per
cmm. One cc of serum obtained thirty days after the onset of the disease
was mixed with 0.1 cc of active virus and injected intracerebrally into a monkey.
Protection was complete, the, animal showing no evidence of poliomyelitis.
The above series of tests of the ability of normal, of suspected,
and of certain poliomyelitic sera to protect monkeys against in-
fection with the virus of poliomyelitis, needs very little comment.
Judging from the results obtained, the test cannot be considered to
give specific evidence as to whether a given individual has or has
not suffered from a previous attack of poliomyelitis. Of the six
normal sera tested, but two failed to protect animals against in-
fection, whereas four gave complete protection, and three of the
latter protected against such large doses of the virus as 0.3 of a cubic
centimeter. It is, however, fair to say that in two of the instances
in which protection was obtained, the individuals had been con-
stantly exposed to infection with the virus of poliomyelitis, and
during the time of exposure had suffered from an indefinite illness
which might possibly be attributed to an abortive attack of polio-
myelitis. In three instances serum was obtained from persons
suspected of having poliomyelitis. Of these sera, one protected a
monkey against experimental infection and two failed to protect.
Two of these individuals were carefully observed in the hospital
Digitized by
Google
18 A CUnical Study of Acute Poliomyelitis.
and both presented fairly definite clinical evidence of the disease.
The serum of one of the latter protected a monkey against infection,
and the serum of the one that failed to protect was obtained to«
early (seven days after the onset of the disease) for the immune
bodies to be present in the serum in sufficient concentration. Four
sera obtained from individuals who had had a characteristic polio-
myelitis protected animals against fatal infection in every instance.
One serum which was obtained thirty years after the occurrence of
the disease, protected against both o.i of a cubic centimeter and 0.3
of a cubic centimeter of active virus. From these results it would
seem that in typical instances of poliomyelitis, the serum of the in-
fected individual probably always contains, after a certain interval,
protective bodies, and that these substances persist for a very long
time. In spite of the irregular results obtained with normal sera
and the sera of suspected individuals, one cannot as yet rule out the
specificity of the test. In the case of apparently normal persons
whose serum protects, the possibility must always be borne in mind
that he may previously have suffered from an unrecognized, abor-
tive attack of the disease. The accurate determination of the pro-
tective value of normal serum can probably be made only by a study
of sera from very young individuals in whom the possibility of pre-
vious infection can certainly be excluded.
PATHOLOGY.
The earliest pathological studies in poliomyelitis were made on
chronic cases of the disease, and the lesions described were the
atrophic scars found in the anterior horns of the cord. With the
development of a better clinical knowledge of the acute stage of the
disease, a new pathology arose which laid especial stress on the
earliest morphological changes in the cord, and explained their rela-
tion to the development of the chronic forms. The work of many
investigators, notably Harbitz and Scheel,*^ Wickman,^* and
Strauss,^ has produced an accurate picture of the anatomical lesions
"Harbitz and Schcel, loc. cit.
"Wickman, Die akute Poliomyelitis, loc cit, p. 13.
"Strauss, The Pathology of Acute Poliomyelitis, Report on the New York
Epidemic of 1907 of the Collective Investigation Committee, New York, 1910.
Digitized by
Google
Francis W. Peahody, Oeorge Draper, and A. B. Dochez. 19
occurring in the brain and spinal cord. The attention of observers
has hitherto been almost wholly centered on the lesions of the central
nervous system, and in consideration of the fact that the most strik-
ing disturbances, both pathological and clinical, are associated with
the nervous system, this is not remarkable. In view, however, of
the very definite and constant changes which are found at autopsy
in other viscera, it is rather strange that they should have been
almost wholly disregarded, and that so little emphasis should have
been put on the fact that acute poliomyelitis is essentially a general
infection. Rissler,^* Strauss, Harbitz and Scheel, and Wickman
have all noted the presence of lesions outside the nervous system,
but have passed them by as having little bearing on the disease. As
a matter of fact, the recognition of acute poliomyelitis as a general
infection has an important bearing both in explaining the clinical
course of the disease and as affecting any possible means of treat-
ing it.
Inasmuch as the lesions of the central nervous system are of chief
importance, they may be considered first. At autopsy the meninges
are usually found to be somewhat edematous and injected. There is
little increase of cerebrospinal fluid. The brain and cord, on section,
have a moist, translucent, edematous appearance, and the gray
matter of the cord is often swollen so that it projects above the level
of the white matter. It is darker than normally in color and is typi-
cally of a grayish pink hue. Not infrequently minute hemorrhages
can be distinguished in the gray or white matter.
The exact path by which the virus enters the body is at present not
definitely known, but there is clinical and experimental evidence
which makes it seem probable that infection frequently gains access
from the upper respiratory tract. It has been shown both anatomi-
cally and experimentally (Flexner^*) that the upper nasal cavities
are in direct communication with the meninges by means of the
lymphatics which pass outward with the filaments of the olfactory
nerve. The view that the virus may enter the body by means of
those lymphatics and thus exert its first effect upon the meninges is
**Rissler, Zur Kcnntniss der Veranderungen des Ncrvcnsystems bd Polio-
myelitis anterior acuta, Nord. med. Ark., 1888, xx, i.
"Flexner, Jour, Am, Med. Assn., loc. cit.
Digitized by
Google
20 A Clinical Study of Acute Poliomyelitis.
strengthened by the anatomical findings. The earliest change which
has been described in the nervous system is hyperemia and the col-
lection of numbers of small mononuclear cells, probably lymjrfio-
cytes, in the perivascular lymph spaces of the blood-vessels of the
leptomeninges. These lymphatic spaces surrounding the vessels are
anatomically processes of the arachnoid spaces, and the lymph in
them is in communication with the cerebrospinal fluid. This first
change, then, is an acute interstitial meningitis, which is not asso-
ciated with fibrin formation or with exudate on the surface of the
meninges. It is usually most marked on the anterior surface of the
spinal cord, and especially in the anterior fissure, from which the
larger vessels enter the cord (figure i) ; but smaller collections of
cells are often found along the meningeal vessels which are situated
over the lateral and posterior aspects of the cord. The blood supply
of the cord is derived from the vessels of the meninges, and with
the advance of the pathological process, this perivascular infiltration
follows along the vessels as they enter the cord from the meninges.
Thus the earliest change that is found in the cord itself, both in
human beings and in the experimentally produced disease, is hyper-
emia and the collection of small round cells in the lymph spaces sur-
rounding the vessels. This cellular exudate forms a sheath ap-
parently completely surrounding the vessels for long stretches
(figure 2), and in many places the cells are so numerous that they
form thick collars which seem to press on the lumen, and thus
exert a mechanical eflFect in obstructing the circulation. While the
cellular exudate is in the outer part of the vessel wall, it is probable
that there is often some effect, either toxic or mechanical, on the
intimal lining of the vessels, for hemorrhages (figure 3), minute
or extensive, are frequent findings, and one of the prominent
features of most cases is the extensive edema. These three factors,
cellular exudate, hemorrhages, and edema, all of them dependent
on vascular changes, may perhaps be regarded as the primary
reaction of the nervous system to the virus of poliomyelitis. The
effects produced on the nerve cells themselves are probably either
dependent on these vascular disturbances or they may be due
to a direct action of the virus. This superior importance of the
vascular system in determining the nervous lesions has for a
Digitized by
Google
Francis W. Pedbody, Oeorge Draper, and A. B. Dochez. 21
long time explained the fact that the cervical and lumbar enlarge-
ments of the cord are most affected, and that the anterior horns of
the gray matter are more involved than the posterior horns or the
white matter. These are, of course, the regions of the spinal cord
to which the blood supply is most abundant. Moreover, the fre-
quency with which lesions are asymmetrical probably depends on
the irregularity with which the vessels supplying the cord are given
off at different levels. It has been suggested that the process by
means of which the vascular lesions affect the nerve cells may be
essentially a mechanical one. It is quite impossible to exclude the
fact that the virus may exert some directly toxic action on these
cells, but in many ways, the clinical and anatomical pictures are
readily explained by the presence of the circulatory disturbance and
of the exudate. On such an hypothesis the damaging effects can
be assumed to result in part from the direct pressure on the nerve
cells of hemorrhages, edema, and exudate. There is also the addi-
tional factor of anemia following the constriction of the blood
vessels by the same mechanism. On account of this pressure and
anemia, the nerve cells degenerate. If the hemorrhage and exudate
are absorbed soon enough, the cells may recover their function. If,
on the other hand, the anemia and pressure have been prolonged or
excessive, the nerve cells go on to complete necrosis. Histological
examination shows nerve cells in all stages of degeneration, from
those with the slightest changes in their protoplasm to others of
which only a granular detritus remains. A most striking picture
is formed by the entrance of polymorphonuclear neurophages into
the necrotic nerve cells (figure 4). A single nerve cell may be in-
vaded by a dozen of these phagocytes, and by means of them ne-
• erotic material is completely disposed of. In more severe lesions,
one sees the hyperemia, the perivascular infiltration, hemorrhages,
edema, and a diffuse cellular infiltration throughout the gray and
white matter, but nerve cells may be completely absent from the
picture. These changes, most prominent In the anterior horns of
the gray matter, are not sharply circumscribed, but are scattered
more or less diffusely through both the gray and white matter of
the cord (figures 5, 6, and 7).
The same sequence of changes, vascular disturbance, and subse-
Digitized by
Google
22 A Clinical Sivdy of Acute Poliomyelitis.
quent degeneration of the nervous elements, is found to a less degree
in the brain, medulla (figure 8), and pons. Hyperemia and a mod-
erate amount of cellular infiltration may be foimd in association
with the vessels of the cerebrum and cerebellum, but lesions exten-
sive enough to produce motor symptoms are exceedingly rare. The
medulla and pons show some slight degree of involvement in most
cases, and one frequently finds in them a marked cellular exudate
and many hemorrhages. It is noteworthy, however, that it is often
extremely difficult to reconcile the clinical S)rmptoms, which are
referable to pontine lesions, with the actual autopsy findings. Cases
which have shown bulbar paralyses in life may fail to show ade-
quate anatomical lesions to account for them, and other cases which
have given clinical evidence of spinal involvement only may show
changes through the pons and medulla.
Of practically constant occurrence are the lesions in the posterior
root ganglia. The histological changes are similar to those that
take place in the cord itself (figures 9 and 10). There is an infiltra-
tion of small round cells in the lymphatic spaces surrounding the
vessels which enter the ganglia from the meninges. This has been
shown experimentally to be the first step in the process. Then fol-
lows a more general, diffuse exudation of cells, degeneration and
necrosis of the nerve cells, and finally the entrance of polymorpho-
nuclear leucocytes into the necrotic cells and removal of the disin-
tegrating cells by neurophages (figure 11). The suggestion has
been made that these lesions in the sensory ganglia may in part
accoimt for the pain which is such a constant feature of the acute
stage of the disease. Another element in the production of pain is
the cellular infiltration which is found along the nerve roots.
The changes which are found in other organs in acute poliomye-
litis are less striking than those in the nervous system, but they