Editorials

Acute viral encephalitis in childhood

Viral infections of the central nervous system are mostly uncommon complications of common systemic viral infections. Remarkably little is known about the factors that determine why a few patients are affected but most are not. While this uncertainty continues the immunisation of the whole child population is necessary for effective prevention.

Acute encephalopathies of viral (or unknown) origin occur most commonly in the first decade of life, with a peak incidence of 1 in 500-1000 infants in their first six months.^{1 2} In practice the diagnosis of encephalitis is often based on an assessment of the clinical features and the exclusion of other possibilities. These include other intracranial infections; Reye's syndrome; hypoxia-ischaemia; haemorrhagic shock and encephalopathy; and metabolic, toxic, vascular, and traumatic disorders. Sometimes the pathogenesis and indeed the diagnosis are left in doubt. Pleocytosis in the cerebrospinal fluid is found in only half of patients with clinically suspected encephalitis, but it helps to narrow the wide differential diagnosis.

No one clinical picture is specific to a single virus or several viruses, and the neurological illness may be triggered by the infecting virus through more than one infective or immune mediated mechanism.^{3 4} Nevertheless, a careful history that takes account of geographical and seasonal factors and a thorough examination may provide important clues to the causal agent.^{1 2}

Encephalitis may present with fever (not invariable) or meningism (with or without altered behaviour), depressed conscious level, seizures, or focal neurological deficits. Electroencephalographic abnormalities are seldom specific but may be helpful in diagnosing herpes simplex encephalitis. Cranial computed tomography often yields normal results or may show non-specific diffuse or focal low attenuation. Magnetic resonance imaging is more sensitive.

Diagnostic tests should include examination and viral culture of samples of the respiratory secretions, a throat swab in viral transport medium, the cerebrospinal fluid, blood, urine, and stool. These specimens should, if possible, be taken within four days of the onset of the illness and sent quickly to the laboratory. Serological investigation is still, however, the mainstay of diagnosis: the proportion of viral infections of the nervous system for which a cause could be established would be greatly increased if antibodies against the relevant group of viruses were sought in appropriate paired serum samples with existing methods.

Before long the polymerase chain reaction (detection of viral antigen by amplification of nucleic acid sequences) is likely to provide a rapid and accurate diagnosis of a wider range of viral infections.^{5 6 7} Measurement of interferon alfa in the cerebrospinal fluid is only moderately sensitive but highly specific for viral infection of the nervous system^{4 6} and may be useful in the acute phase--but the test is not widely available. Positive identification of viral infection in the central nervous system helps to curtail investigation, rationalise treatment, and improve the reliability of prognosis.

Two additional convalescent serological investigations are underused. Firstly, the IgG index is a measure of intrathecal synthesis of IgG derived by calculating the ratios of the concentrations of total IgG and albumin in the cerebrospinal fluid to those in serum. A raised IgG index and the presence of oligoclonal bands suggest that intrathecal synthesis of IgG has been taking place, but the test cannot link the illness to a specific virus. Secondly, however, the IgG ratio specific to a virus compares the ratio of specific antiviral IgG in the cerebrospinal fluid to that in the serum with that of total IgG, and this test can be used to identify infection by a specific virus.^{3 4 5 6 8}

The main viral causes of acute meningitis and encephalitis in Britain are enteroviruses, paramyxoviruses (mumps, measles, respiratory syncytial virus, parainfluenza), herpes viruses, influenza, rubella, and adenoviruses.^{1 2 3 4} Immunologically mediated disease is one variant⁸; classically it occurs as part of a biphasic illness and affects white matter predominantly. In some cases, poorly defined, treatment with steroids may be helpful.⁹ Mycoplasma pneumoniae is also a common cause of encephalitis.^{2 10} Worldwide, rabies and Japanese B encephalitis (an arboviral disease that kills 1000 people a year in China alone) are the most common causes of death from encephalitis.^{3 4}

How, then, should a patient with presumed viral encephalitis be managed? The three immediate steps are to give a high dose of intravenous acyclovir promptly (even in the absence of focal features that suggest herpes virus infection), to search for other treatable causes of an acute encephalopathy, and to protect the child's brain against further insult. Supportive care should include providing nutrition; controlling seizures, the temperature, and electrolyte balance; and treating dysfunction of other organs--a severe neurological deficit in acute encephalitis does not preclude a good recovery. If the cerebral perfusion pressure (the difference between systolic blood pressure and intracranial pressure) is decreased in children the mortality is raised.¹¹ Maintenance of the cerebral perfusion pressure, if necessary by increasing the arterial pressure, seems a logical goal of treatment, though there is little direct evidence that such intervention improves the prognosis.

Clinical features are helpful in predicting the outcome. Young age, low conscious level, abnormal oculocephalic responses, a high ratio of albumin in cerebrospinal fluid to that in serum, and laboratory evidence of infection of the central nervous system are pointers to a poor prognosis.^{4 10} The case fatality rate was 34% among 338 children under 3 years of age with acute encephalopathies enrolled in the national childhood encephalopathy study and was similar whether or not encephalitis was specifically diagnosed. At follow up 10 years later almost half the survivors had motor dysfunction and educational dysfunction; well over a third had neurological dysfunction; and almost a fifth had behavioural, self care, and sensory dysfunction. Mortality and morbidity were even higher in children with multisystem disease (who hardly ever had pleocytosis in the cerebrospinal fluid).¹² In a study of 462 Finnish children under 17 years of age the mortality and serious morbidity of encephalitis were 2.8% and 6.7%, with a sevenfold and 12-fold increased risk of these outcomes after M pneumoniae encephalitis and herpes simplex virus encephalitis respectively.¹⁰

Acute childhood encephalopathies make a substantial contribution to chronic neurological handicap, and the impact on individual families (frequently exacerbated by diagnostic uncertainty) may be devastating. The recent launch of a patient support group (helpline telephone number 01751 432 369) for those affected by these illnesses is therefore welcome.

Consultant paediatric neurologist Department of Paediatric Neurology, Southampton General Hospital, Southampton SO16 6YD

Colin Kennedy 

1. Wang D, Bortolussi R. Acute viral infection of the CNS in children: an eight year review. Can Med Assoc J 1981;125:585-9. [Abstract]
2. Koskiniemi M, Rautonen J, Lehtokoski-Lehtiniemi E, Vaheri A. Epidemiology of encephalitis in children: a 20 year survey. Ann Neurol 1991;29:492-7. [Medline]
3. Johnson RT. Viral infections of the nervous system. New York: Raven Press, 1982.
4. Kennedy CR. Acute viral infections excluding herpes simplex, rabies and HIV. In: Lambert HP, ed. Infections of the central nervous system. London: Edward Arnold, 1991.
5. Aurelius E, Johansson B, Skoldenberg B, Staland A, Forsgren M. Rapid diagnosis of herpes simplex encephalitis by nested polymerase chain reaction assay of cerebrospinal fluid. Lancet 1991;337: 189-92.
6. Rozenberg F, Lebon P. Amplification and characterisation of herpes virus DNA in cerebrospinal fluid from patients with acute encephalitis. J Clin Microbiol 1991;29:2412-7. [Abstract/Free Full Text]
7. Rotbart HA. Diagnosis of enteroviral meningitis with the polymerase chain reaction. J Pediatr 1990;117:85-9. [Medline]
8. Johnson RT, Griffin DE, Hirsch BL, Wolinsky JS, Roedenbeck S, Lindo de Soriano I, et al. Measles encephalomyelitis--clinical and immunological studies. N Engl J Med 1984;310:137-41. [Abstract]
9. Pasternak JF, De Vivo DC, Prensky AL. Steroid responsive encephalomyelitis in childhood. Neurology 1980;30:481-6. [Abstract/Free Full Text]
10. Rautonen J, Koskiniemi M, Vaheri A. Prognostic factors in childhood acute encephalitis. Paediatr Infect Dis J 1991;10:441-6. [Medline]
11. Goitein K, Tamir I. Cerebral perfusion pressure in central nervous system infections of infancy and childhood. J Pediatr 1983;103:40-3. [Medline]
12. Madge N, Diamond J, Miller D, Ross E, McManus C, Wadsworth J, et al. The National Childhood Encephalopathy Study: a 10-year follow-up. Dev Med Child Neurol 1993;35:(suppl 68):26,51.