The anticipated eradication of poliovirus emphasizes the need to identify other enteroviral causes of severe central nervous system disease. Enterovirus 68 has been implicated only in cases of respiratory illness. We therefore report a case of fatal meningomyeloencephalitis caused by enterovirus 68 in a 5-year-old boy, which required neuropathology, microbiology, and molecular techniques to diagnose.

The patient was a previously healthy 5-year-old boy with up-to-date immunizations who initially sought care for headache, low-grade fever, sore throat, and unilateral neck tenderness in the fall of 2008. He had no tick exposure or travel outside of New Hampshire. A number of classmates reportedly had coldlike illnesses. During the following 2 days, he experienced myalgia and progressive weakness of his arms (right arm to a greater extent than left) and a change in the timbre of his voice. Four days after onset he visited his primary care provider who believed he had a nonspecific viral illness. He had a white blood cell count of 10 900/µL (80% neutrophils, 5% bands, and 12% lymphocytes). Serum electrolyte and liver function test results were normal. The erythrocyte sedimentation rate was 16 mm/h. In the next few hours the patient developed bowel and bladder incontinence and the inability to walk. Later that evening he was found in bed, apneic and unresponsive, with intermittent tonic posturing of the upper extremities. He had a cardiac arrest en route to the hospital, and on admission to the pediatric intensive care unit he remained unresponsive and had a temperature of 32.5°C, an arterial blood pH of 6.5, and a Pco2 of 102 mm Hg. A computed tomography scan of the brain showed diffuse cerebral edema with tentorial herniation. A chest x-ray showed a hazy opacity of the right lower lobe. As clinical criteria for brain death were met, life support was withdrawn and a full autopsy performed.

Cerebrospinal fluid (CSF) obtained by cisternal puncture showed slight xanthochromia; a nucleated cell count of 2094/μL (54% lymphocytes); glucose, 2 mg/dL; and protein, 368 mg/dL. Tissue and CSF were sent for diagnostic studies at the Dartmouth-Hitchcock Medical Center (Lebanon, New Hampshire), the New Hampshire Public Health Laboratory (Concord, New Hampshire), and the Centers for Disease Control and Prevention (CDC), Polio and Picornavirus Branch (Atlanta, Georgia). Findings from the bacterial cultures of the frontal cortex, CSF, lung, stool, and urine were all negative. Results of viral cultures (herpes simplex virus, cytomegalovirus, varicella zoster virus, and enterovirus) of the frontal lobe, lung, and CSF were negative. Screening of CSF for West Nile and Eastern equine encephalitis viruses by reverse transcriptase–polymerase chain reaction (RT-PCR) yielded negative results at the state laboratory. The CDC identified enterovirus 68 in the patient's CSF but not the frontal lobe, using VP1-specific RT-PCR and sequencing.1 An attempt to amplify enterovirus 68 from formalin-fixed lung tissue was unsuccessful owing to RNA degradation. Immunoglobulin (Ig) levels (IgG, 505 mg/dL; IgA, 44 mg/dL; and IgM, 62 mg/dL) were all normal for the patient′s age.

The patient's lungs were 3 times the normal weight and diffusely firm. Microscopic sections showed acute mixed pneumonia with hemorrhage (Figure, A). The spleen was twice the normal weight and florid follicular hyperplasia was observed microscopically. Sections of the thymus and bone marrow showed normal immune architecture.

A, Lung shows severe acute inflammation consistent with acute pneumonia. B, Anterior spinal cord shows diffuse parenchymal lymphocytic host response with neuronophagia of motor neurons by infiltrating lymphocytes. C, Anterior spinal cord characterizes the parenchymal lymphocytes as T cells. D, Anterior spinal cord demonstrates the predominantly perivascular location of B cells. E, Anterior spinal cord demonstrates the lack of activation of the apoptotic pathway by motor neurons. F, Anterior spinal cord demonstrates the T-cell cytolytic protein located on the motor neurons (hematoxylin-eosin, original magnifications ×100 [A] and ×200 [B]; anti-CD3 antibody, original magnification ×100 [C]; anti-CD20 antibody, original magnification ×100 [D]; anti–caspase-3 antibody, original magnification ×40 [E]; anti-perforin antibody, original magnification ×200 [F]).

A, Lung shows severe acute inflammation consistent with acute pneumonia. B, Anterior spinal cord shows diffuse parenchymal lymphocytic host response with neuronophagia of motor neurons by infiltrating lymphocytes. C, Anterior spinal cord characterizes the parenchymal lymphocytes as T cells. D, Anterior spinal cord demonstrates the predominantly perivascular location of B cells. E, Anterior spinal cord demonstrates the lack of activation of the apoptotic pathway by motor neurons. F, Anterior spinal cord demonstrates the T-cell cytolytic protein located on the motor neurons (hematoxylin-eosin, original magnifications ×100 [A] and ×200 [B]; anti-CD3 antibody, original magnification ×100 [C]; anti-CD20 antibody, original magnification ×100 [D]; anti–caspase-3 antibody, original magnification ×40 [E]; anti-perforin antibody, original magnification ×200 [F]).

Close modal

Postmortem gross examination of the brain demonstrated profound edema and normal-appearing leptomeninges. The brainstem was grossly unremarkable, although multiple red-brown discolorations averaging 7 mm in greatest diameter were identified in the dentate nucleus bilaterally. Microscopic examination of the hippocampus revealed hypereosinophilic neurons with pyknotic nuclei, consistent with hypoxic-ischemic injury. Microscopy of the meninges, cerebellum, midbrain, pons, medulla, and cervical cord demonstrated extensive lymphocytic meningomyelitis and encephalitis, characterized by prominent neuronophagia in motor nuclei (Figure, B).

Immunochemical stains for CD3 in the spinal cord highlighted a diffuse parenchymal infiltration of T lymphocytes in and around motor nuclei (Figure, C). The CD20 stain in the spinal cord identified B lymphocytes predominantly localized to the perivascular areas (Figure, D). Antibody staining for caspase-3, a component of apoptotic cell death, was negative in anterior spinal cord α-motor neurons (Figure, E). Most α-motor neurons were positive for perforin, the cytolytic protein used by CD8 T cells and natural killer cells to cause cell death (Figure, F). Gomori methenamine silver and Gram stain results were negative in lung sections. Immunohistochemical studies of lung and spinal cord with an available, but unvalidated, anti–enterovirus 68 horse polyclonal antiserum were not definitive.

This child presented with acute neurologic impairment, flaccid paralysis, and pneumonia. The patient's lumbar puncture was consistent with aseptic meningitis when correcting for postmortem decrease in glucose.2 Pertinent negative findings include negative findings for bacterial cultures, viral cultures, special stains for fungi and bacteria and negative PCR results for 2 recognized causes of encephalitis in New Hampshire (West Nile and Eastern equine encephalitis viruses). Identification and sequencing of enterovirus 68 in the spinal fluid, combined with the characteristic histopathologic pattern of enteroviral central nervous system infection, established this virus as the cause of the fatal illness.3 

The identification of the pathogen as enterovirus 68 is based on the nucleotide sequence of the viral DNA present in the patient, specifically the portion of the viral genome region encoding the VP1 capsid protein. The sequence in this region has been shown to correlate with enterovirus serotype, as determined by antigenic methods.4 In this case, a portion of VP1 (∼350 nucleotides) was amplified by seminested RT-PCR and the sequence was compared with sequences in a database representing all known enterovirus serotypes, using a previously described method.1 The VP1 sequence was unique (not identical to that of any other strain in the laboratory) and it was 85.0% identical to the EV68 prototype strain, indicating a successful identification (strains of the same type share ≥75% nucleotide identity in this region).

The human enteroviruses are classified into 4 species in the genus Enterovirus on the basis of their genetic relationships.5 Human enterovirus A (HEV-A) and HEV-C each contain some of the coxsackie A viruses and several of the more recently numbered enteroviruses, and polioviruses are classified within HEV-C. The coxsackie B viruses, echoviruses, and most of the numbered enteroviruses are in HEV-B, while HEV-D includes EV68, EV70, and EV94. Enterovirus 68 has not been previously reported as a cause of neurologic disease. Although enteroviruses are a well-recognized cause of aseptic meningitis,3 and can cause disseminated disease in the newborn6 and focal encephalitis,7 the identification of a new enterovirus agent as a cause of serious neurologic disease is noteworthy. The presenting symptoms of this case, with asymmetric paralysis, and the histologic findings of neuronophagia of motor nuclei in the anterior horn cells are similar to disease caused by the 3 poliovirus serotypes.

The pathologic findings in the brainstem and the severity of disease are most analogous to viral meningitis, acute motor neuron disease, and brainstem encephalitis that complicate infections with enterovirus 71, a species A enterovirus.8 In fact, the sudden cardiovascular collapse observed in our patient with enterovirus 68 brainstem encephalitis is strikingly similar to a rapidly fatal course—complicating some cases of acute enterovirus 71 encephalitis—that has been attributed to rapid onset of neurogenic pulmonary edema.911 In this case, the postmortem lung findings demonstrated pneumonia rather than pulmonary edema. Previously, enterovirus 68 has been isolated primarily from the respiratory tract. It has been described as having phenotypic characteristics of a rhinovirus, as it is more acid-sensitive and thermolabile than other enterovirus serotypes.12,13 Although we failed to identify enterovirus 68 in the lungs of this patient, we used a suboptimal cell mix for replication (R-Mix, Diagnostic Hybrids, Athens, Ohio) and higher than optimal temperature for enterovirus 68 detection in tissue culture. The radiologic and pathologic evidence of pneumonia without a bacterial etiology suggests that systemic spread of enterovirus 68 from a respiratory source may have been a component of the pathogenesis of this case. Most efforts at identification of causes of polio-like illnesses are based on enteric shedding of viruses, such that enterovirus 68–associated illnesses, if primarily replicating in the respiratory tract, might be missed. The normal immunoglobulin levels and unremarkable medical history make it unlikely that this patient had a B-cell defect that predisposed him to more serious enterovirus disease.14 Alternative causes of viral encephalitis were ruled out by molecular diagnostic techniques and culture and histopathologic findings.

The immunopathogenesis of enteroviral infection is not completely understood. What has been learned comes from the study of myocarditis.15 A competent immune system with intact T-cell function is necessary for viral clearance. In general, B cells inhibit viral replication and geographically contain the virus with the help of macrophages, whereas T cells enter into these areas of viral infection, resulting in tissue damage and cell death. In this case report, abundant T cells were seen infiltrating motor nuclei of the brain, presumably leading to the observed cell death of motor neurons (Figure, C). To determine whether motor neurons were dying as a result of apoptosis versus necrosis, staining for caspase-3 (a proapoptotic protein) and perforin (a pore-forming protein that causes necrosis) were performed. Motor neurons showed no staining with anti–caspase-3 but stained strongly with anti-perforin, consistent with the proposed cytotoxic role played by T cells in this patient's infection (Figure, E and F).

Enterovirus 68 appears to be a rare cause of neurologic disease. The CDC Picornavirus Laboratory has identified only 1 other enterovirus 68 from CSF in the past 10 years. Nevertheless, identification of enterovirus 68 and other novel enterovirus neuropathogens will increase our awareness of their clinical role and pathogenesis and help establish their true incidence.

We thank Jason Pettus, MD, for his thorough postmortem examination and David Beck, HTL(ASCP) for his expertise in immunohistochemical staining.

1
Nix
,
W.
,
M.
Oberste
, and
M.
Pallansch
.
Sensitive, seminested PCR amplification of VP1 sequences for direct identification of all enterovirus serotypes from original clinical specimens.
J Clin Microbiol
2006
.
44
(
8
):
2698
2704
.
2
Karkela
,
J. T.
Critical evaluation of postmortem changes in human autopsy cisternal fluid—enzymes, electrolytes, acid-base balance, glucose and glycolysis, free amino acids and ammonia: correlation to total brain ischemia.
J Forensic Sci
1993
.
38
(
3
):
603
616
.
3
Ellison
,
D.
,
S.
Love
,
L.
Chimelli
,
B.
Harding
,
J.
Lowe
, and
H.
Vinters
.
Acute viral infections.
In:
.
Neuropathology: A Reference Text of CNS Pathology. 2nd ed
.
Oxford
England: Elsevier
.
2004
.
273
275
.
4
Oberste
,
M. S.
and
M. A.
Pallansch
.
Enterovirus molecular detection and typing.
Rev Med Microbiol
2005
.
16
:
163
171
.
5
Stanway
,
G.
,
F.
Brown
,
P.
Christian
, et al
.
Picornaviridae.
In:
Fauquet
,
C. M.
,
M. A.
Mayo
,
J.
Maniloff
,
U.
Desselberger
, and
L. A.
Ball
.
eds.
Virus Taxonomy: Eighth Report of the International Committee on the Taxonomy of Viruses
.
Amsterdam, the Netherlands
Elsevier Academic Press
.
2005
.
757
778
.
6
Modlin
,
J. F.
Fatal echovirus 11 disease in premature neonates.
Pediatrics
1980
.
66
(
5
):
775
780
.
7
Modlin
,
J. F.
,
R.
Dagan
,
L. E.
Berlin
,
D. M.
Virshup
,
R. H.
Yolken
, and
M.
Menegus
.
Focal encephalitis with enterovirus infections.
Pediatrics
1991
.
88
(
4
):
841
845
.
8
Alexander
Jr,
J. P.
,
L.
Baden
,
M. A.
Pallansch
, and
L. J.
Anderson
.
Enterovirus 71 infections and neurologic disease—United States, 1977–1991.
J Infect Dis
1994
.
169
(
4
):
905
908
.
9
Ho
,
M.
,
E. R.
Chen
,
K. H.
Hsu
, et al
.
An epidemic of enterovirus 71 infection in Taiwan.
N Engl J Med
1999
.
341
(
13
):
929
935
.
10
Huang
,
C-C.
,
C-C.
Liu
,
Y-C.
Chang
,
C-Y.
Chen
,
S-T.
Wang
, and
T-F.
Yeh
.
Neurologic complications in children with enterovirus 71 infection.
N Engl J Med
1999
.
341
(
13
):
936
942
.
11
Wang
,
S. M.
,
C. C.
Liu
,
H. W.
Tseng
, et al
.
Clinical spectrum of enterovirus 71 infection in children in southern Taiwan, with an emphasis on neurological complications.
Clin Infect Dis
1999
.
29
(
1
):
184
190
.
12
Oberste
,
M. S.
,
K.
Maher
,
D.
Schnurr
, et al
.
Enterovirus 68 is associated with respiratory illness and shares biological features with both the enteroviruses and the rhinoviruses.
J Gen Virol
2004
.
85
(
9
):
2577
2584
.
13
Blomqvist
,
S.
,
C.
Savolainen
,
L.
Råman
,
M.
Roivainen
, and
T.
Hovi
.
Human rhinovirus 87 and enterovirus 68 represent a unique serotype with enterovirus and rhinovirus features.
J Clin Microbiol
2002
.
40
(
11
):
4218
4223
.
14
McKinney
,
R. E.
,
S. L.
Katz
, and
C. M.
Wilfert
.
Chronic enteroviral meningoencephalitis in agammaglobulinemic patients.
Rev Infect Dis
1987
.
9
(
2
):
334
356
.
15
Cooper
,
L. T.
Myocarditis.
N Engl J Med
2009
.
360
(
15
):
1526
1538
.

Author notes

From the Departments of Pathology (Drs Kreuter, Schwartzman, and Rhodes), Medicine (Dr Barnes), and Pediatrics (Drs McCarthy, Modlin, and Wright), Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire; and the Department of Virology, Centers for Disease Control and Prevention, Atlanta, Georgia (Dr Oberste).

The authors have no relevant financial interest in the products or companies described in this article.