Journal of Neuropathology and Experimental Neurology
Copyright q 2001 by the American Association of Neuropathologists
Vol. 60, No. 7
July, 2001
pp. 696 704
Influenza RNA not Detected in Archival Brain Tissues from Acute Encephalitis Lethargica
Cases or in Postencephalitic Parkinson Cases
SHERMAN MCCALL, MD, JAMES M. HENRY, MD, ANN H. REID, MA,
AND
JEFFERY K. TAUBENBERGER, MD, PHD
Abstract. Encephalitis lethargica (EL) was a mysterious epidemic, temporally associated with the 1918 Spanish influenza
pandemic. Numerous symptoms characterized this disease, including headache, diplopia, fever, fatal coma, delirium, oculogyric
crisis, lethargy, catatonia, and psychiatric symptoms. Many patients who initially recovered subsequently developed profound,
chronic parkinsonism. The etiologic association of influenza with EL is controversial. Five acute EL autopsies and more than
70 postencephalitic parkinsonian autopsies were available in the Armed Forces Institute of Pathology (AFIP) tissue repository.
Two of these 5 acute EL cases had histopathologic changes consistent with that diagnosis. The remaining 3 cases were
classified as possible acute EL cases as the autopsy material was insufficient for detailed histopathologic examination. RNA
lysates were prepared from 29 CNS autopsy tissue blocks from the 5 acute cases and 9 lysates from blocks containing
substantia nigra from 2 postencephalitic cases. RNA recovery was assessed by amplification of beta-2-microglobulin mRNA
and 65% of the tissue blocks contained amplifiable RNA. Reverse transcription-polymerase chain reaction (RT-PCR) for
influenza matrix and nucleoprotein genes was negative in all cases. Thus, it is unlikely that the 1918 influenza virus was
neurotropic and directly responsible for the outbreak of EL.
Key Words:
Encephalitis lethargica; Influenza; Postencephalitic parkinsonism.
INTRODUCTION
Encephalitis lethargica (EL), or von Economo’s encephalitis, has remained a mysterious disease entity since
its initial description in 1917 (1, 2). A serious and often
lethal form of encephalitis, it caused a widespread outbreak throughout Europe and North America during the
years 1918–1925 (3–5). One of the difficulties in explaining the outbreak was the profusion of clinical signs and
symptoms ascribed to EL. Three well-described variants
were observed: the somnolent-ophthalmoplegic form, the
hyperkinetic form, and the amyostatic-akinetic form (5,
6). Nevertheless, most contemporary observers felt the
large clinical overlap made these a single disease entity.
Overall case mortality during the epidemic was approximately 40%. A second remarkable feature of this encephalitic syndrome was the complication of postencephalitic parkinsonism (PEP) (5–7). Unlike idiopathic
Parkinson disease (IP), which most commonly occurs in
individuals over the age of 50, PEP developed in EL
survivors of all ages including young adults (8–10). Although few new acute EL cases were reported after 1925,
the number of PEP cases continued to accumulate into
the 1930s. Since then, despite occasional isolated case
reports (11–15), there has been no conclusive recurrence
of the disease. A detailed survey of the EL outbreak can
be found in the companion review published in this issue
(2).
From the Division of Molecular Pathology, Department of Cellular
Pathology and Genetics (SM, AHR, JKT), and Department of Neuropathology (JMH), Armed Forces Institute of Pathology, Washington,
DC.
Correspondence to: Jeffery K. Taubenberger, MD, PhD, Department
of Cellular Pathology, Armed Forces Institute of Pathology, 1413 Research Blvd., Bldg. 101, Room 1057D, Rockville, MD 20850-3125.
In the fall of 1918, an influenza pandemic of unprecedented proportions swept the globe in a few months (16,
17). The so-called ‘Spanish’ influenza pandemic was
caused by an H1N1 influenza A virus. At least 500 million people were symptomatically infected (18), and
globally up to 40 million people died of acute influenza
and the complications of pneumonia (16). The pandemic
progressed in epidemiologically recognizable waves with
a first wave in the spring of 1918, which was characterized by an infectious but not particularly lethal course.
The second wave emerged in late August, causing the
main global outbreak from September 1918 to February
1919. It was characterized by exceptional severity and
high mortality, especially among young adults. Third and
fourth waves of the pandemic (1919–1921) caused successively less extensive illness as immunity to the 1918
strain apparently became more widespread.
While contemporary observers of the EL epidemic and
the 1918 influenza pandemic felt confident that these represented distinct and independent disease entities (4, 5),
the 2 outbreaks have become linked in the modern medical literature (7, 19), although a causal relationship has
never been documented (20). The most popular current
theory postulates that the 1918 influenza virus was directly neurotropic, i.e. capable of replicating within the
central nervous system (CNS) (20, 21). This theory is
based on the unusual virulence of the 1918 pandemic, the
description of neurotropism in a mouse-adapted influenza
strain (22–24), and the coincident temporal relationship
of the 1918 influenza and EL outbreaks. However, since
both outbreaks occurred over 80 yr ago and EL has not
recurred in epidemic form, resolution of this medical dilemma was thought unlikely. Moreover, the hypothesis
that 1918 influenza virus replication could be maintained
696
INFLUENZA RNA NOT DETECTED IN ENCEPHALITIS LETHARGICA
chronically in EL has been supported by evidence of persistent infection by a replication-defective measles virus
in subacute sclerosing panencephalitis (SSPE) (20).
Recently however, it became possible to examine the
genetic structure of the 1918 influenza virus directly using RNA isolated from archival fixed and frozen lung
tissue samples from victims of the 1918 pandemic (25–
27). Sequence analysis has demonstrated that the 2 surface protein-encoding genes, while having features consistent with mammalian adaptation, were probably
derived from an avian influenza source shortly before the
pandemic (25–27). Neither protein has known mutations
that would allow the virus to replicate outside the respiratory tree. These findings are supported by pathological
examination of autopsy tissues of 1918 flu victims, which
show respiratory disease as their primary pathology (28–
30) and an absence of CNS abnormalities.
Using the same techniques that led to the demonstration of influenza RNA in archival 1918 autopsy lung
samples (25–27), a study was undertaken to examine
brain tissues from acute EL and chronic PEP fatalities for
the presence of influenza RNA. Influenza RNA could not
be isolated in any of the cases examined, suggesting that
the 1918 influenza virus did not directly infect the brain
and was not the direct cause of EL.
MATERIALS AND METHODS
Because encephalitis lethargica ostensibly represented a new
disease, it was diagnosed under a plethora of names. The
Armed Forces Institute of Pathology (AFIP) tissue repository
was searched for 14 known synonyms yielding 29 cases coded
with diagnoses consistent with EL. CNS tissue was available
from 5 patients who died in a primary attack or acute recurrence. Two of these cases had histopathologic changes consistent with the diagnosis of acute EL. The remaining 3 cases had
insufficient tissue for detailed histologic analysis. These cases
are included as possible acute EL cases. The repository was
also searched for 7 synonyms of postencephalitic parkinsonism
yielding 78 cases. To test for influenza virus in a chronic infectious form, 2 representative PEP cases were tested for which
tissue blocks containing substantia nigra were available.
Case Histories
Although the available medical and autopsy records are
sometimes incomplete, a synopsis of patient medical histories
follows. CSF changes were reported in about 50% of encephalitis lethargica cases, including an increase in opening pressure, a mildly increased protein, and slight lymphocytosis. The
gross appearance, culture, Wassermann reaction, and Fehling’s
reduction test were reported to be normal in EL (31). (Colloidal
gold is a test for CSF protein (32) and the Wassermann reaction
is a complement fixation test for syphilis (33).)
Acute EL Case 1: A 39-yr-old Asian male in Manila exhibited
symptoms which began on December 29, 1922 with pain, abdominal hyperreflexia, and spasm. He had tachycardia, dry
mouth, and a temperature of 99.88F. He was restless, picked at
697
his bedclothes and was in constant motion. He twitched spasmodically, had muttering speech, a coarse intentional and resting tremor, and an unsteady gait with a tendency to fall backward. From January 13–20, 1923, decreased abdominal reflexes
were noted. He experienced bouts of irrationality during which
he walked around nude or with a sheet over his head, climbed
into bed with other patients and hit them with pillows. He
washed his clothes and then climbed into bed with wet pajamas.
He was not aware of his behavior, but answered questions appropriately. On January 30 he began having urinary retention
with up to 1,330 ml on catheterization, but his urinalysis was
normal. On February 1 he died after becoming febrile (axillary
temperature of 1078F) and stuporous. His blood leukocyte count
was 7,200 with a differential of 30 lymphocytes, 4 monocytes,
63 neutrophils, and 4 bands. His Wassermann reaction was negative at autopsy and the autopsy report states that the brain had
an ‘‘irregular and patchy perivascular infiltrate with some generalized lymphocytic infiltration. Ganglion cells of medulla
show fairly advanced degeneration.’’ The cause of death was
listed as ‘‘encephalitis lethargica, myoclonic type.’’
Acute EL Case 2: An archival case of acute EL was submitted
previously at the request of the AFIP by Jefferson Medical College, Philadelphia. The date and clinical history of the case are
unknown. While no case records are available for this case, the
file diagnosis of acute EL is supported by histologic examination (see Results).
Acute EL Case 3 (Possible): A 45-yr-old Caucasian male became ill on February 8, 1923 and died February 19, 1923 with
a diagnosis of ‘‘lethargic encephalitis.’’ At autopsy the brain
was grossly unremarkable but, according to the autopsy report,
it showed a ‘‘diffuse lymphocytic infiltrate in caudate, choroid
plexus, medulla and cord most marked in section through 4th
ventricle. Ganglion cells in medulla show rather advanced chromatolysis of individual cells, with atrophy and loss of nuclear
staining. There are numerous hemorrhages from small and medium sized vessels throughout the section of the medulla with
edema most marked near the floor of the 4th ventricle.’’
Acute EL Case 4 (Possible): A 40-yr-old Caucasian male who
had influenza in 1918. Six months later he collapsed and was
blind for weeks before recovering with diplopia. He then developed rousable somnolence for 8 months. On recovery he had
a shuffling walk, with left foot drag and amnesia. Abdominal
colic followed and the patient became enema dependent. In January 1926 he lost use of both arms, followed by throat paralysis
in August and aphasia in November, when he died of bronchopneumonia. At the time of death the CSF was clear, with a sugar
of 86.9 mg, 12 cells per mm3, and negative Wassermann, globulin, and colloidal gold tests. At autopsy the CNS was grossly
normal, although no histologic report was provided.
Acute EL Case 5 (Possible): A 19-yr-old Caucasian male Alcatraz prisoner who had influenza in 1918. On June 6, 1927 he
was dizzy and had a seizure without fever. After 2 days his
dizziness returned with right temporal headache and no focal
findings. On June 10 he was stuporous, but rousable, with slight
nuchal rigidity and an equivocal Babinski, right hemiparesis,
and sluggish reflexes throughout. He was normotensive and had
a temperature of 1028F. The leukocyte count was 13,000 with
75% neutrophils, falling in succeeding days to 6,800, with a
negative blood culture. The CSF was clear with 48 cells per
mm3, a sugar of 85.4 mg, and negative globulin, Wassermann,
J Neuropathol Exp Neurol, Vol 60, July, 2001
698
McCALL ET AL
and colloidal gold tests. His eyes did not converge and corneal
reflexes were intact with no evidence of ptosis. There was a
left hemianesthesia and areflexia, an equivocal right Babinski
reflex, and hyperreflexia, and pseudoclonus on passive motion.
Abdominal, epigastric, cremasteric and left patellar reflexes
were absent. On June 13 he died with tachycardia and a fever
of 1058F. There had been no polio or influenza outbreaks on
Alcatraz for the previous 2 yr. His medical and trauma history
was negative.
At autopsy the brain and cord weighed 1,550 g with markedly dilated vessels. The autopsy report shows that there was
no suppurative process, but a subarachnoid hemorrhage was
present over the precentral and paracentral gyri. Petechial hemorrhages and a 1-cm hemorrhage were present in the precentral
and paracentral regions in the gray and white matter. There was
vascular engorgement and perivascular infiltration throughout
brain. In the right cerebrum there was edema and necrosis with
infiltrates consisting chiefly of lymphocytes, plasma cells,
monocytes and activated macrophages, but few neutrophils.
Perivascular and some nonvascular hemorrhage was present.
PEP Case 1: A 60-yr-old Caucasian male was hospitalized
since 1926 with tremor of the mandible, legs, and hands. There
was permanent flexion of the fingers, elbows and knees. His
mentation was good, but speech was almost unintelligible. He
died in 1950 with a diagnosis of ‘‘chronic encephalitis lethargica,’’ although no immediate cause of death was documented.
The autopsy report states that the CSF was watery and clear
and the gross examination of the dura and brain was normal.
Grossly, the substantia nigra was very pale and rarified, reflecting severe loss of melanin. The neurons were markedly decreased in number with considerable gliosis.
PEP Case 2: A 71-yr-old Caucasian male who contracted
influenza in World War I subsequently developed a progressive
tremor after the war mandating retirement in 1937 based on a
diagnosis of postencephalitic parkinsonism. He was hospitalized intermittently until 1951, after which time he required fulltime nursing care. Bilateral chemopallidectomy performed in
1957 provided no improvement. He had an Achilles tendonotomy in 1961 for severe contracture of the legs. He died in
1968 from bronchopneumonia.
The autopsy report shows that the fresh brain weighed 1,270
g and 1,310 g after fixation. Gyral atrophy was especially
marked in the frontal lobes. Near the right putamen there was
a well-circumscribed, partly calcified mass measuring 2.2 3 1.7
3 4 cm. In the left internal capsule and periventricular portion
of brain there was a similar mass with dimensions of 1.1 3 0.8
3 3 cm, extending into the substantia nigra. Both were due to
the chemopallidectomy. There was no abnormality in the pons
or medulla oblongata. There was a gray soft area, 2 cm in
diameter, in the right cerebellum near the dentate nucleus and
attached cortex. Representative sections were taken from the
right putamen, left internal capsule, substantia nigra, and dentate nucleus. Right and left sections from the pons, medulla
oblongata, and cervical medulla were also submitted.
Negative Control Case: A 25-yr-old caucasian male who died
on May 29, 1919 of a dural meningioma with massive cyst
formation and pressure atrophy of the brain. This case corresponds temporally to the end of the third wave of the influenza
pandemic and is included as a negative control (Fig. 5).
J Neuropathol Exp Neurol, Vol 60, July, 2001
RNA Isolation and RT-PCR
Tissue blocks were deparaffinized, RNA extracts prepared,
and RT-PCR reactions were performed utilizing published procedures that have been used successfully to amplify viral RNA
from 1918 influenza pneumonia victim lung tissue (27, 34).
Three primer sets for the matrix (M) and nucleoprotein (NP)
gene segments were designed from consensus sequences of
1930s H1N1 viruses. Primer sequences are as follows: 59AYGGGDGAYCCAAAYAACATGGA-39 (forward) and 59YCARCTGGCWAGTGCACCAG-39 (reverse) amplified a 120
bp fragment of the M1 gene (nucleotides 287–406 of the M
gene segment as aligned to A/Puerto Rico/8/34 (H1N1),
GenBank accession number V01099). 59-ACAGATTGCTGATTCCCAGC-39 (forward) and 59-CATAGCCTTAGCYGTAGTGC-39 (reverse) amplified a 112 bp fragment of the M1
gene (nucleotides 481–592 of the M gene segment as aligned
A/Puerto Rico/8/34 (H1N1), GenBank accession number
V01099). 59-ATGGAYCCCAGRATGTGYTC-39 (forward) and
59-ARYTCCAWNAYCATTGTYCC-39 (reverse) amplified a
104 bp fragment of the NP gene (nucleotides 520–623 of the
NP gene as aligned to A/Puerto Rico/8/34 (H1N1), GenBank
accession number M38279). Positive control RNA included A/
Puerto Rico/8/34 (H1N1) and the 1918 influenza strain A/Brevig Mission/1/18 (H1N1). As a control for RNA quality, RTPCR for a fragment of the b2-microglobulin gene was performed as previously described (27, 34).
RESULTS
Histopathology
The histopathologic changes characterizing the acute
cases of von Economo’s encephalitis lethargica include
diffuse parenchymal congestion and scattered, perivascular ring hemorrhages involving small, thin-walled vessels. There are focal round cell infiltrates in the leptomeninges with random perivascular round cell infiltrates
in the parenchyma, consisting of lymphocytes admixed
with plasma cells (Fig. 1). Variable degrees of parenchymal round cell infiltrates, neuronophagia, and gliosis are
noted, with no evidence of viral inclusion bodies. Isolated
foci of white matter necrosis are present in acute EL case
2. These changes are particularly prominent in the brainstem, although occasional foci are also present in the cerebral cortex.
Sections available for histopathologic examination in
acute EL case 1 include pons, medulla, and spinal cord.
Sections of the pons include the locus ceruleus and a
midpontine section through the abducens and facial nerve
nuclei. The parenchyma contains multiple foci of dense,
perivascular round cell infiltrates within the VirchowRobin spaces of small vessels, consisting predominantly
of lymphocytes. Similar perivascular infiltrates involve
the locus ceruleus, although there is no evidence of neuronophagia or loss of pigment. Sections of midpons contain intact neurons within the abducens and facial nerve
nuclei. Sections of the medulla and spinal cord contain
similar inflammatory changes, including scattered foci of
INFLUENZA RNA NOT DETECTED IN ENCEPHALITIS LETHARGICA
699
Fig. 1. Perivascular round cell cuffing characterizing acute encephalitis lethargica (H&E), particularly prominent in the brainstem, including pigmented and nonpigmented nuclei (original magnification 3100).
Fig. 2. Globose neurofibrillary tangles (Bodian) distributed throughout the CNS, with emphasis on the brainstem, including
pigmented and nonpigmented nuclei and the colliculi (original magnification 3600).
J Neuropathol Exp Neurol, Vol 60, July, 2001
700
McCALL ET AL
leptomeningeal round cell infiltrates with superficial extension into the Virchow-Robin spaces. There is marked
vascular congestion with occasional small, perivascular
ring hemorrhages.
Histopathologic examination of acute EL case 2 reveals
a dorsal portion of rostral brainstem. There is intense perivascular inflammation within the parenchyma, consisting
of dense perivascular round cell infiltrates in the VirchowRobin spaces, with a preponderance of lymphocytes and
admixed plasma cells. There are similar inflammatory cells
within the parenchymal stroma with evidence of tissue injury, ranging from rarefaction and gliosis to frank necrosis
of white matter with macrophage activity. Neuronophagia
is absent within the adjacent gray matter. Congestion with
small, perivascular ring hemorrhages is observed. These
changes reflect a more severe degree of parenchymal injury than acute EL case 1.
Two chronic cases of PEP show severe neuronal loss
within the substantia nigra (SN). The remaining neurons
of SN show prominent argyrophilic, intracytoplasmic
neurofibrillary tangles (NFT), predominantly of the globose type, in addition to occasional flame shaped NFT
(Fig. 2). Similar NFT are present within the remaining
brainstem nuclei, being particularly prominent in the occulomotor nucleus and the region of the colliculi, all of
which are characterized by tau-immunoreactivity. There
is severe, diffuse astrogliosis throughout the parenchyma
of the brainstem, including SN and the region of the colliculi. Although bona fide senile plaques are not present,
occasional structures resembling senile plaques are observed (Fig. 3), consistent in appearance with nonamyloid astrocytic plaques (35, 36). Many astrocytes contain
small argyrophilic cytoplasmic inclusions, characterized
by immunoreactivity for tau, in the absence of affinity
for alpha-synuclein (a-syn) (Fig. 4), consistent in appearance with glial fibrillary tangles (GFT) of the astrocytic type, also designated as astrocytic fibrillary tangles
(37–39). There is also evidence of extensive, diffuse
swelling of neurites, reflecting the presence of diffuse
axonal injury (DAI) (Fig. 5). (A detailed report of these
observations will be made in a subsequent publication.)
Molecular Examination
A total of 29 lysates from 22 CNS tissue blocks from
5 acute EL cases and 9 blocks of SN from 2 PEP cases
were tested by RT-PCR for influenza RNA. RNA lysates
were also tested for expression of gene b2-microglobulin
mRNA (expressed in all nucleated cells) in order to evaluate RNA recovery. Amplifiable b2-microglobulin
mRNA was observed from 100% of the cases and 65%
of the total tissue blocks (Fig. 6), including the negative
control endothelioma case from 1919.
Influenza mRNA from the 2 most plentiful viral proteins were chosen as RT-PCR targets: matrix (3,000 copies per virus) and nucleoprotein (1,000 copies) (40). Nucleoprotein appears within 2 hours of infection (41),
J Neuropathol Exp Neurol, Vol 60, July, 2001
while matrix protein appears within 5 hr of infection (42)
and continues to be synthesized as a major component of
budding virions. RT-PCR for 2 fragments of the matrix
1 gene and for a fragment of the nucleoprotein gene was
negative in all EL and PEP samples (Fig. 6). Positive
controls included influenza A/Puerto Rico/8/34 (H1N1)
and A/Brevig Mission/1/18 (H1N1) from 1918 influenza
lung tissue (26). The negative controls included the control case from 1919 and samples without added RNA
template to control for PCR contamination. Negative control lanes were uniformly negative for influenza RNA.
DISCUSSION
While contemporary observers doubted a connection
between the 1918 influenza virus and the epidemic of EL
(4, 5, 43), later medical thought has tended to link them
causally (7, 19). Several groups have linked these 2 outbreaks to a single etiologic agent, predominantly because
they overlapped in time and because of the unusual virulence of the 1918 influenza (7, 19, 21). The EL outbreak, which emerged around 1916, prior to the first
wave of the influenza pandemic, resulted in thousands of
cases in Europe and North America before ending around
1925. Available records suggest that there were approximately 25,000–40,000 clinical cases of EL in the United
States (2). Of these, 7,700 people died of acute EL, while
15,000 people died of complications of PEP. In contrast,
the influenza pandemic of 1918–1919 caused clinical disease in 25%–33% of the US population, representing approximately 30 million cases, with 675,000 deaths occurring from October 1918 to March 1919 alone (16, 17).
Since the 1918 influenza strain was not isolated at the
time, and no convincing pathogen was isolated from the
brains of EL victims, subsequent studies on these 2 diseases have been limited to available archival tissue (44–
48). Recently, it became feasible to isolate influenza RNA
from frozen and formalin-fixed, paraffin-embedded autopsy lung tissue. Using this material, complete genomic
sequences of the influenza hemagglutinin, neuraminidase,
and nonstructural genes have been obtained. None of the
genes possesses mutations that are known to allow other
influenza viruses to become pantropic or neurotropic (25–
27).
Several previous studies have examined EL and PEP
brain tissue for influenza RNA or protein antigens (7). In
2 studies from the 1970s, Gamboa et al (46, 47) employed a direct immunofluorescence technique on frozen
brain sections of 6 PEP cases using antisera derived from
several different influenza A and B strains, measles virus,
and herpes simplex virus. Antisera were derived from the
neurotropic mouse-adapted influenza strains, A/WSN/33
(H1N1) and A/NWS/33 (H1N1), which are lab strains
derived from influenza A/WS/33 (H1N1) by serial inoculation in ferrets, egg culture, and mice brains (22, 23).
Positive staining using antisera from these 2 virus strains
INFLUENZA RNA NOT DETECTED IN ENCEPHALITIS LETHARGICA
701
Fig. 3. Nonamyloid, tau-positive astrocytic plaque comprised of thickened, abnormal astrocytic processes, superficially resembling senile plaques of Alzheimer disease (original magnification 3640).
Fig. 4. Glial fibrillary tangles of the astrocytic type (astrocytic tangles) presenting as tau-positive inclusions within the
cytoplasm of astrocytes (original magnification 3640).
J Neuropathol Exp Neurol, Vol 60, July, 2001
702
McCALL ET AL
Fig. 5. Diffuse fragmentation and swelling of axons (Bielschowsky), particularly prominent within the brainstem, reflecting
nontraumatic diffuse axonal injury (original magnification 3600).
Fig. 6. Representative RT-PCR amplification results: 14 of the 29 total acute EL CNS blocks are shown. RNA was isolated
and RT-PCR performed for influenza RNA and b2 microglobulin (b2m) mRNA as a control for RNA integrity. A band using
the b2m primer set is observed in 12 of the 14 preparations shown (lanes 8 and 10 are negative). None of the preparations show
a band using the influenza primer sets, including the set for the matrix 1 gene shown here. Lane 15 is the negative control lane
containing no template RNA and lane 16 is the positive control.
was reported in the substantia nigra and hypothalamus of
6 postencephalitic Parkinson patients. However, they also
reported positive staining in individual cases with antisera
from an H2N2 influenza A strain, A/Swine/Iowa/30
J Neuropathol Exp Neurol, Vol 60, July, 2001
(H1N1), and herpes simplex virus, suggesting that these
results may not be specific.
In more recent studies, several techniques have failed
to identify influenza antigens or RNA in EL or PEP brain
INFLUENZA RNA NOT DETECTED IN ENCEPHALITIS LETHARGICA
tissues. Two archival cases (1 EL, 1 PEP) were negative
for influenza A and B, herpes, rubella, cytomegalovirus,
measles, and mumps by immunohistochemical analysis
(44, 49). In situ RT-PCR for influenza A in 7 PEP cases
was also negative (48).
The current study used the sensitive technique of RTPCR to examine archival brain tissues from acute EL and
PEP fatalities. Three primer sets derived from the matrix
and nucleoprotein genes were negative in all samples.
Control amplification using RT-PCR primers for the b2microglobulin gene was positive in tissue lysates from all
cases, with an overall positivity rate of 65% of RNA
lysates examined. These results corroborate the above
studies, revealing no evidence of influenza in EL or PEP
brains. The same technique has yielded complete gene
sequences from RNA isolated from the lungs of 1918
influenza pneumonia victims (25–27) and morbillivirus
RNA in marine mammal epizootics (50, 51). In both cases, amplifiable RNA could be recovered from tissue that
was too poorly preserved for detailed histologic analysis.
EL remains a mysterious disease with no definite etiology. Although no mechanism underlying chronic persistence of influenza virus in CNS tissue has been described, the fact that the 1918 influenza virus showed
unusual properties (21), and that persistent measles virus
infection is observed in SSPE, has led investigators to
speculate, by analogy, that the 1918 influenza virus may
have been associated with chronic CNS persistence in
cases of EL and PEP (20). A careful review of the available contemporary literature, as well as clinical and epidemiologic data, however suggest that EL-PEP was unlikely to have been caused by influenza. The results of
the current study and several previous experimental studies support this hypothesis. Also supporting this hypothesis is the direct sequence analysis of the 1918 influenza
virus. Neither of the surface proteins, hemagglutinin and
neuraminidase, possess mutations which would allow the
1918 influenza virus to become pantropic or neurotropic.
While the EL epidemic behaved as an infectious disease
(without evidence of person-to-person transmission), no
other likely candidate organism has been described. The
pathology of acute EL is nonspecific and is very similar
to that seen in other viral encephalitides, such as Japanese
encephalitis virus. Unlike other known mosquito-borne
arboviral encephalitic diseases however, EL occurred during the winter months in the northern hemisphere. While
other viral encephalitides occasionally cause either transient or long-term postencephalitic parkinsonian syndromes, they are not generally associated with the development of secondary parkinsonism (7). The absence
of virus in patients with a clinical course as short as 7
days and the confluence of other data indicates that EL
was not caused directly by the 1918 influenza virus. The
pathology, as well as the clinical and epidemiologic data
suggest that EL was caused by an unknown infectious
703
agent. The theoretical possibility remains that some secondary effect of influenza infection, e.g. autoimmune,
was responsible. Future work using archival EL tissue
may provide clues to the etiology of this enigmatic disease.
ACKNOWLEDGMENTS
The authors thank Dr. David Morens and Professor Kurt Jellinger for
helpful discussion. The opinions or assertions contained herein are the
private views of the authors and are not to be construed as official or
as reflecting the views of the Department of the Army or the Department
of Defense. This is US government work; there are no restrictions on
its use.
REFERENCES
1. von Economo C. Encephalitis lethargica. Wiener Klinische Wochenschrift 1917;30:581–85
2. Reid AH, McCall S, Henry JM, Taubenberger JK. Experimenting
on the past: The enigma of von Economo’s encephalitis lethargica.
J Neuropathol Exp Neurol 2000; 60:663–70
3. Tilney F, Howe HS. Epidemic Encephalitis. New York: Paul B.
Hoeber, 1920
4. Hall AJ. Epidemic encephalitis. Bristol: John Wright and Sons,
LTD., 1924
5. von Economo C. Encephalitis lethargica: Its sequelae and treatment.
London: Oxford University Press, 1931
6. Bing R. Zur Frage des ‘‘Parkinsonismus’’ als Folgezustand der Encephalitis lethargica. Schweizer Medizinische Wochenschrift 1921;
51:4–7
7. Casals J, Elizan TS, Yahr MD. Postencephalitic parkinsonism—A
review. J Neural Transm 1998;105:645–76
8. Clarke C. Mortality from Parkinson’s disease in England and Wales
1921–89. J Neurol, Neurosurg, Psychi 1992;56:690–93
9. Dimsdale H. Changes in the parkinsonian syndrome in the twentieth
century. Quart J Med 1946;15:155–70
10. Duvoisin RC. The cause of Parkinson’s disease.In: Marsden CD,
Fahn S, eds. Neurology 2: Movement disorders. London: Butterworth Scientific, 1982
11. Pilz P. Gibt es noch eine Encephalitis lethargica Economo? Wiener
Medizinische Wochenschrift 1978;128:762–63
12. Gulmann NC, Pedersen HE. Parkinsonisme efter akut encefalopati—Encephalitis lethargica—en forsvunden sygdom? Ugeskr Laeger 1980;42:960–61
13. Greenough A, Davis JA. Encephalitis lethargica: Mystery of the
past or undiagnosed disease of the present? Lancet 1983;1:922–23
14. Kapadia F, Grant IS. Encephalitis lethargica. Intensive Care Med
1990;16:338–39
15. Barletta L, et al. Encephalitis lethargica. Nervenarzt 1995;66:781–4
16. Reid AH, Taubenberger JK. The 1918 flu and other influenza pandemics: ‘‘Over there’’ and back again. Lab Invest 1999;79:95–101
17. Crosby A. America’s Forgotten Pandemic. Cambridge: Cambridge
University Press, 1989
18. Laidlaw P. Epidemic Influenza: A virus disease. Lancet 1935;228:
1118–24
19. Ravenholt R, Foege W. 1918 influenza, encephalitis lethargica, parkinsonism. Lancet 1982;2:860–64
20. Schlesinger RW, Husak PJ, Bradshaw GL, Panayotov PP. Mechanisms involved in natural and experimental neuropathogenicity of
influenza viruses: Evidence and speculation. Advances in Virus Research 1998;50:289–379
21. Maurizi C. Influenza caused epidemic encephalitis (encephalitis
lethargica): The circumstantial evidence and a challenge to the naysayers. Medical Hypothesis 1989;28:139–42
J Neuropathol Exp Neurol, Vol 60, July, 2001
704
McCALL ET AL
22. Stuart-Harris CH. A neurotropic strain of influenza virus. Lancet
1939:497–99
23. Francis T, Moore A. Study of the neurotropic tendency in strains
of the virus of epidemic influenza. J Exp Med 1940;72:717–28
24. Taubenberger JK. Influenza virus hemagglutinin cleavage into HA1,
HA2: No laughing matter. Proc Natl Acad Sci 1998;95:9713–15
25. Reid AH, Fanning TG, Janczewski TA, Taubenberger JK. Characterization of the 1918 ‘‘Spanish’’ influenza virus neuraminidase
gene. Proc Natl Acad Sci U S A 2000;97:6785–90
26. Reid AH, Fanning TG, Hultin JV, Taubenberger JK. Origin and
evolution of the 1918 ‘‘spanish’’ influenza virus hemagglutinin
gene. Proc Natl Acad Sci USA 1999;96:1651–56
27. Taubenberger JK, Reid AH, Krafft AE, Bijwaard KE, Fanning TG.
Initial genetic characterization of the 1918 ‘‘Spanish’’ influenza virus. Science 1997;275:1793–96
28. Wolbach SB. Comments on the pathology and bacteriology of fatal
influenza cases, as observed at Camp Devens, Mass. Johns Hopkins
Hospital Bulletin 1919;30:104–9
29. Winternitz MC, Wason IM, McNamara FP. The pathology of influenza. New Haven: Yale University Press, 1920
30. LeCount ER. The pathologic anatomy of influenzal bronchopneumonia. JAMA 1918;72:650–52
31. Smith H. Epidemic encephalitis (encephalitis lethargica, Nona). Report of studies conducted in the United States. Pub Health Reports
1921;36:207–42
32. Wallach J. Interpretation of diagnostic tests. Boston: Little, Brown
and Company, 1986
33. Basmajian J. Stedman’s medical dictionary. Baltimore: Williams &
Wilkins, 1982
34. Krafft AE, Duncan BW, Bijwaard KE, Taubenberger JK, Lichy JH.
Optimization of the isolation and amplification of rna from formalin-fixed, paraffin-embedded tissue: The Armed Forces Institute
of Pathology experience and literature review. Mol Diagn 1997;2:
217–30
35. Hallervorden J. Anatomische Untersuchungen zur Pathogenese des
postencephalitischen Parkinsonismus. Deutsche Zeitschrift für Nervenheilkunde 1935;136:68–77
36. Feany MB, Dickson DW. Widespread cytoskeletal pathology characterizes corticobasal degeneration. Am J Pathol 1995;146:1388–96
37. Haraguchi T, et al. An autopsy case of postencephalitic parkinsonism of von Economo type: Some new observations concerning neurofibrillary tangles and astrocytic tangles. Neuropathology 2000;20:
143–48
J Neuropathol Exp Neurol, Vol 60, July, 2001
38. Ikeda K, Akiyama H, Kondo H. Anti-tau-positive glial fibrillary
tangles in the brain of postencephalitic parkinsonism of Economo
type. Neurosci Lett 1993;162:176–78
39. Chin SS, Goldman JE. Glial inclusions in CNS degenerative diseases. J Neuropathol Exp Neurol 1996;55:499–508
40. Skehel J, Schild G. Polypeptide composition of influenza A viruses.
Virology 1971;44:396–408
41. Liu O. Studies on influenza infection in ferrets by means of fluorescein-labelled antibody. J Exp Med 1955;101:677
42. Oxford J, Schild G. Immunological studies with influenza matrix
protein. In: Barry R, Mahy B, eds. Immunological studies with
influenza matrix protein. New York: Academic Press, 1975
43. Jordan E. The influenza epidemic of 1918: Encephalitis and influenza. JAMA 1927;89:1603–6
44. Elizan T, Casals J, Swash M. No viral antigens detected in brain
tissue from a case of acute encephalitis lethargica and another case
of post-encephalitic Parkinsonism. J Neurol Neurosurg Psychiatry
1989;52:800–801
45. Elizan TS, Casals J. Astrogliosis in von Economo’s and postencephalitic Parkinson’s diseases supports probable viral etiology. J
Neurol Sci 1991;105:131–34
46. Gamboa ET, Wolf A, Yahr MD, Barden H, Hsu CC, Duffy PE,
Harter DH. Influenza A virus as a possible cause of postencephalitic
Parkinsonism. Trans Am Neurol Assoc 1973;98:177–80
47. Gamboa E, Wolf W, Yahr M, et al. Influenza virus antigen in postencephalitic parkinsonism brain. Arch Neurol 1974;31:228–32
48. Isaacson SH, Ascher DM, Gibbs CJ, Jr., Gajdusek DC, Yahr MD.
In situ RT-PCR of influenza viral RNA. Society for Neuroscience.
Abstr Soc Neurosci 1995;21:1272
49. Esiri M, Swash M. Absence of herpes simplex virus antigen in brain
in encephalitis lethargica. J Neurol Neurosurg Psychiatry 1984;47:
1049–50
50. Krafft A, Lichy JH, Lipscomb TP, Klaunberg BA, Kennedy S, Taubenberger JK. Postmortem diagnosis of morbillivirus infection in
bottlenose dolphins (Tursiops truncatus) in the Atlantic and Gulf
of Mexico epizootics by polymerase chain reaction-based assay. J
Wildl Dis 1995;31:410–15
51. Taubenberger JK, Tsai M, Krafft AE, et al. Two morbilliviruses
implicated in bottlenose dolphin epizootics. Emerg Infect Dis 1996;
2:213–16
Received December 13, 2000
Revision received April 2, 2001
Accepted April 6, 2001