Estimates are that >100 children in Japan have died recently of influenza type A-associated encephalopathy and almost the same number of children have severe sequelae every year.1, 2 Most of the reported patients have been young children. Influenza type A-associated encephalopathy is typically associated with sudden onset of high fever, severe convulsions and rapidly progressive coma and death within 2 or 3 days.1 These patients have generally been treated only with supportive measures, although many of them received acyclovir intravenously. Amantadine has been used to treat influenza type A infections, mainly for respiratory complications, such as bronchitis and pneumonia.3 During the 1995 to 1996 epidemic caused by influenza A/Texas/36/91(H1N1), we treated a severe case of influenza type A-associated encephalopathy with amantadine. Case report. A 2-year-old boy was well until 1 day before admission to our hospital when he developed a high fever. The following morning he had a generalized tonic-clonic convulsion and was transferred to our hospital. On admission the patient was comatose, with involuntary choreoathetoid movements. The pupils were large and symmetric with sluggish reaction to light. A cerebrospinal fluid specimen contained only one cell, and the protein and glucose concentrations were normal. Electroencephalography revealed high voltage semirhythmic delta waves. Computed tomography of the brain yielded normal findings. Laboratory studies showed no liver or biochemical abnormalities. The clinical impression was acute encephalopathy. The patient was given acyclovir, dexamethasone and anticonvulsants, and he was given assisted ventilation. Starting on the 3rd day of illness, he was also given amantadine, 100 mg/day (∼6 mg/kg) through a stomach tube for 7 days. On the 5th day of illness he began to breathe spontaneously without assisted ventilation. On the 8th day of illness he was extubated. On the 15th day of illness he could sit up in bed. He was discharged 30 days after admission with no residual symptoms. The plasma amantadine concentration 1 day after discontinuing treatment was 220 ng/ml. Influenza A(H1N1) virus was isolated from the throat swab on admission. Serum hemagglutination inhibition antibody titers against A/Texas/36/91(H1N1) increased from <1:8 to 1:128 between the acute and convalescent phase. Additional serologic tests failed to demonstrate any other viruses, including complement-fixation titers of herpes simplex virus, influenza A and B, adenovirus, respiratory syncytial virus and Mycoplasma pneumoniae, plus the hemagglutination inhibition titers of influenza A (H3N2) and B and parainfluenza virus types 1 to 3 were negative. Discussion. Our patient showed a typical clinical course of influenza type A-associated encephalopathy recently reported in Japan.1, 2 Strangely influenza type A-associated encephalopathy has received little attention in the US and European countries.4 Diagnosis and classification as acute encephalopathy of unknown origin during the winter season may be usual in these countries because most pediatricians do not regard the influenza virus as a causative agent of acute encephalopathy except in Reye's syndrome. Involvement of the central nervous system (CNS) in influenza may be either by direct invasion by the virus or by a hypersensitivity response.5 In our patient the influenza type A virus may have directly invaded the CNS through viremia, because the patient became comatose soon after developing a febrile illness. The effectiveness of amantadine is unknown in this case. However, if influenza virus is present in the CNS, amantadine might be effective. Amantadine provides complete bioavailability orally, and the cerebrospinal fluid concentrations are about one-half the concurrent plasma values.3 If the plasma concentration exceeds 400 ng/ml, amantadine might be effective for treating influenza A encephalopathy because the amantadine concentrations that inhibit 50% plaque formation by the influenza A virus range from 200 to 400 ng/ml.3 Prompt intravenous administration of acyclovir is recommended to treat encephalitis despite the absence of focal features that suggest herpesvirus infection.6 According to recent experiences in Japan, influenza type A-associated encephalopathy is much more common than generally appreciated. We suggest that, during influenza type A epidemics, children with presumed viral encephalitis be treated with amantadine, in addition to acyclovir, especially when influenza type A infection is demonstrated by rapid antigen detection tests.7 Norio Sugaya, M.D. Masaru Miura, M.D. Department of Pediatrics; Nippon Kokan Hospital; Kawasaki (NS) Department of Pediatrics; Yokohama Municipal Citizen's Hospital; Yokohama (MM) Japan