Viral encephalitis is a febrile illness with evidence of brain parenchymal dysfunction manifested by an altered state of consciousness and/or objective signs of neurologic dysfunction as a result of a viral infection. This review discusses the most appropriate tests to establish the diagnosis of the more common nonzoonotic viral encephalitides.
Herpes Simplex Viruses (HSVs)
HSVs are responsible for 10–20% of all cases of identifiable viral encephalitides in the United States (US). Although HSV-1 and -2 can cause encephalitis in neonates, almost all cases of HSV encephalitis in older children are the result of HSV-1 infection. In neonates, clinical findings include bulging fontanel, poor feeding, fever, temperature instability, irritability, lethargy and frequently seizures. Older children may have fever, altered level of consciousness, behavioral changes and focal neurologic deficits associated with temporal lobe involvement.1–3
Cerebrospinal fluid (CSF) analysis in the neonate generally demonstrates <100 white blood cells (WBC)/mm3 with a mononuclear predominance and elevated protein. The CSF profile in older children consists of a lymphocytic predominant pleocytosis of 10–500 cells/mm3. In some cases, the CSF may be hemorrhagic. Magnetic resonance imaging (MRI) is more sensitive than computed tomography (CT) for the detection of early brain changes that typically localize to temporal lobes. Electroencephalographic (EEG) examination reveals focal changes arising from the temporal lobes consisting of spike and slow-wave activity and periodic lateralized epileptiform discharges (PLEDs).1–3
HSV can be isolated from the CSF by culture in only 40% of neonatal HSV encephalitis cases and 2% of older children and adults with HSV encephalitis. Consequently detection of HSV DNA in the CSF using polymerase chain reaction (PCR) has become the test of choice for establishing the diagnosis of HSV encephalitis.4 Data from the Collaborative Antiviral Study Group found the sensitivity and specificity of PCR to be 94 and 98%, respectively, compared with brain biopsy-confirmed cases.5 Overall the sensitivity and specificity of PCR detection in cases of neonatal HSV encephalitis range from 75 to 100% and 71 to 100%, respectively.4 Based on other studies of PCR diagnosis of HSV encephalitis, the sensitivity of PCR in the diagnosis of nonneonatal HSV encephalitis is 97%.4 Detection of type-specific HSV antibodies in serum is possible3; however, their utility in the diagnosis of HSV encephalitis is limited.
Enteroviruses (EV)
EV cause 10–20% of identifiable viral encephalitides. Group A coxsackieviruses are the most common subgroup isolated from these cases. The presentation of EV encephalitis consists of fever and mental status changes that include disorientation, focal seizures and occasionally coma.6 Severe rhombencephalitis has been reported with EV 71 infection.7
The CSF typically demonstrates a normal cell count or mild monocytic pleocytosis with a minimally elevated protein and normal glucose concentration. Neuroimaging is normal in the majority, but diffuse or focal lesions on CT or MRI have been reported. In EV 71-associated encephalitis, MRI may demonstrate lesions in the brain stem, especially the tegmentum. Focal disturbances may be noted on EEG.6,7
The test of choice for establishing the diagnosis of EV encephalitis is EV genome detection in the CSF using reverse transcription (RT)-PCR. Although no published studies evaluating the sensitivity and specificity of RT-PCR in patients with EV encephalitis exist, RT-PCR has been demonstrated to have a sensitivity and specificity of 86–100% and 92–100%, respectively, for the diagnosis of EV meningitis.4
Influenza Viruses
Influenza A and B viruses can cause encephalitis. In Japan, influenza A encephalitis occurs primarily in children younger than 5 years of age, with case-fatality rates as high as 37%.8 This high severity has not been reported in the US. In the US, children typically present with fever and upper respiratory symptoms followed within 2 days by the onset of neurologic symptoms typically consisting of decreased responsiveness, speech difficulties, motor difficulties and seizures.8,9
In the majority of cases, the CSF analysis is normal. Case series from the US have generally failed to identify abnormalities on neuroimaging. Reported neuroimaging abnormalities from Japan include cerebral edema and bilateral symmetric thalamic, cerebral and/or pontine lesions. EEG findings have included diffuse slowing as well as focal disturbances such as PLEDs.8–10
No confirmatory test exists for the diagnosis of influenza encephalitis. Although reports of detection of influenza virus genome in the CSF using RT-PCR exist,8 these are exceptions rather than the rule. Isolation of influenza virus from the CSF via cell culture has been rare. Therefore a presumptive causal relationship between influenza virus infection and encephalitis is generally established by detecting virus (by cell culture or antigen detection) from the nasopharynx of a patient with neurologic symptoms consistent with encephalitis.
Nonsimplex Herpesviridae
The nonsimplex Herpesviridae [varicella-zoster (VZV), Epstein-Barr virus (EBV), human herpesvirus 6 (HHV-6) and cytomegalovirus] are infrequent causes of encephalitis in immunocompetent children. VZV encephalitis and cerebellar ataxia occur in <1% of infected healthy children.11 In the former, neurologic signs and symptoms begin ∼1 week after the rash develops. In some cases, no rash precedes central nervous system (CNS) findings. These include fever, headache, emesis, seizures, ataxia, decreased level of consciousness, altered reflexes, hyper- or hypotonia, paresis and extensor plantar reflexes.11
In VZV encephalitis, CSF analysis usually reveals a lymphocytic pleocytosis of <100 cells/mm3, elevated protein and normal glucose. The opening pressure may be increased. EEG findings consist primarily of diffuse slow wave activity. Neuroimaging studies may demonstrate cerebral edema or findings consistent with demyelination.11 Because VZV has been rarely cultured from the CSF, the diagnostic modality of choice is PCR detection of the VZV genome in CSF.12
Cerebellar ataxia may present from days before the onset of rash to 2 weeks after its appearance.11 Emesis, headache, lethargy, meningismus and nuchal rigidity may accompany the ataxia. The CSF is generally normal but may demonstrate mild lymphocytic pleocytosis and increased protein concentration.
EBV has been associated with encephalitis, meningoencephalitis and cerebellar ataxia. Although no unique symptom complex accompanies EBV encephalitis, several reports have linked it with metamorphosia (Alice in Wonderland syndrome). PCR is the test of choice for attempting to detect EBV in the CSF.12
In the immunocompetent host, although HHV-6 genome has been detected in the CSF of infants undergoing primary HHV-6 infection, encephalitis is an uncommon event.12 As with other Herpesviridae, PCR detection of viral genome in the CSF should be considered the method of choice for attempting to establish a diagnosis of HHV-6-associated encephalitis.
Mumps Virus
The current outbreak of mumps in the Midwest13 has drawn attention to the encephalitic potential of this virus. Before development of the mumps vaccine, it was the most common etiology among reportable cases of encephalitis. Children with mumps encephalitis were primarily male with a mean age of ∼7 years. Patients presented with fever and commonly had parotitis before the onset of encephalitis. Neurologic signs and symptoms consisted of impaired locomotion and balance, seizures, psychiatric disorders and sensory disturbances.14
The CSF reveals <500 WBC/mm3 in association with a normal protein concentration and, unlike most viral CNS infections, a slightly depressed glucose concentration.14 Findings on neuroimaging include cerebral edema, transverse myelitis and demyelinated areas including the thalamus, caudate and cerebellum.15 In the majority, EEG findings consist of moderate to severe slowing.14
The diagnosis of mumps encephalitis may be established by detection of virus or viral genome from the CSF or documentation of serologic titers in association with encephalitis. Mumps is easily cultured from CSF, saliva or throat swabs. RT-PCR detection is at least equivalent to cell culture for identification of mumps from the CSF and is significantly faster.16 When cell culture or PCR are unsuccessful in establishing the diagnosis, detection of viral specific IgM or a 4-fold rise in serologic titers may be useful. Caution must be taken in interpreting these results because antibodies to mumps may cross-react with those to parainfluenza viruses.
REFERENCES
1. Kimberlin D. Herpes simplex virus infections in neonates. …
Semin Pediatr Infect Dis. 2005;16:271–281.
2. Kimberlin D. Herpes simplex virus infections. …
Semin Pediatr Infect Dis. 2003;14:83–89.
3. Kimberlin DW. Neonatal herpes simplex infection.
Clin Microbiol Rev. 2004;17:1–13.
4. Romero JR, Kimberlin DW. Molecular diagnosis of viral infections. …
Clin Lab Med. 2003;23:843–865.
5. Lakeman FD, Whitley RJ. Diagnosis of herpes simplex
encephalitis. …
J Infect Dis. 1995;172:857–863.
6. Modlin JF, et al. Focal
encephalitis with enterovirus infections.
Pediatrics. 1991;88:841–845.
7. Huang CC, et al. Neurologic complications in children. …
N Engl J Med. 1999;341:936–942.
8. Newland JG, et al.
Encephalitis associated with influenza B. …
Clin Infect Dis. 2003;36:e87–e95.
9. Maricich SM, et al. Neurologic complications associated with influenza A. …
Pediatrics. 2004;114:e626–e633.
10. Morishima T, et al.
Encephalitis and encephalopathy. …
Clin Infect Dis. 2002;35:512–517.
11. Gnann JW Jr. Varicella-zoster virus: atypical presentations.…
J Infect Dis. 2002;186(Suppl 1):S91–S98.
12. DeBiasi RL, Tyler KL. Molecular methods. …
Clin Microbiol Rev. 2004;17:903–925.
13. CDC. Mump epidemic.…
MMWR. 2006;55:366–368.
14. Koskiniemi M, et al. Clinical appearance and outcome in mumps
encephalitis. …
Acta Paediatr Scand. 1983;72:603–609.
15. Tarr RW, et al. MRI of mumps
encephalitis. …
Pediatr Radiol. 1987;17:59–62.
16. Poggio GP, et al. Nested PCR for rapid detection of mumps virus. …
J Clin Microbiol. 2000;38:274–278.