Acute Necrotizing Encephalopathy of Childhood Associated With Influenza Type B Virus Infection in a 3-Year-Old Girl.

Navigation: Main page

Author: Huang, Shih-Ming1Chen, Chu-Chin1 childdoctor@hotiriail.com.Chiu, Pao-Chin1Cheng, Ming-Fang1Lai, Ping

Acute necrotizing encephalopathy of childhood represents a novel entity of acute encephalophathy, predominantly affecting infants and young children living in Taiwan and Japan. It manifests with symptoms of coma, convulsions, and hyperpyrexia after 2 to 4 days of respiratory tract infections in previously healthy children. The hallmark of acute necrotizing encephalopathy of childhood consists of multifocal and symmetric brain lesions affecting the bilateral thalami, brainstem tegmentum, cerebral periventricular white matter, or cerebellar medulla. The etiology and pathogenesis of this kind of acute encephalopathy remain unknown, and there is no specific therapy or prevention. The prognosis is usually poor, and less than 10% of patients recover completely. We report a 3-year-old previously healthy girl presenting with acute necrotizing encephalopathy of childhood associated with influenza type B virus infection, which resulted in severe neurologic sequelae. We also review the current knowledge of the clinical, neuroimaging, and pathologic aspects of acute necrotizing encephalopathy of childhood. (J Child Neural 2004; 19:64-67).

In 1995, Mizuguchi et al reported a possibly novel subset of acute encephalopathy predominantly affecting Japanese and Taiwanese children characterized by the presence of symmetric thalamic lesions along with abnormal liver function, termed acute necrotizing encephalopathy of childhood.&sup1,2; Upper respiratory infection often precedes the encephalopathy, and influenza type A virus is the most common pathogen found in these cases. There are no symptoms specific to acute necrotizing encephalopathy of childhood, and the onset of the first sign of brain dysfunction, including vomiting, seizure, and coma, ranges from 0.5 to 4 days.&sup2,3; The role of influenza virus infections in the pathogenesis of acute necrotizing encephalopathy of childhood remains unknown. Although influenza virus is demonstrated in the throats of patients, cerebrospinal fluid specimens usually test negative by virus culture and polymerase chain reaction, and no perivascular infiltration of inflammatory cells could be found in the acute-stage autopsy of the acute necrotizing encephalopathy of childhood. It could differentiate acute necrotizing encephalopathy of childhood from infected encephalitis or acute disseminated encephaloniyelitis.&sup1-4; Acute necrotizing encephalopathy of childhood might be initiated by influenza infection in endothelial cells in selected areas of the brain, such as the thalamus, followed by breakdown of the blood brain barrier and the extravasation of blood.[sup5] We report a 3-year-old previously healthy girl presenting with acute necrotizing encephalopathy of childhood associated with influenza type B virus infection, which caused severe neurologic sequelae. We also review the current knowledge of the clinical, neuroimaging, and pathologic aspects of acute necrotizing encephalopathy of childhood.

In February 2003, a 3-year-old previously healthy girl developed a flulike illness 3 days before admission characterized by cough, rhinorrhea, and fever up to 40°C. On the day of admission, she experienced febrile generalized tonic-clonic seizures. She was sent to a local hospital, where diazepam was administered intramuscularly, and the seizures ceased. She was referred to our hospital 7 hours later because of persistent impairment of consciousness. On arrival at our pediatric intensive care unit, she was comatose, combined with bilateral pinpoint pupils without light reflexes. Urine retention was also found simultaneously. The initial vital signs showed a heart rate of 180/minute, respiratory rate of 45/minute, central temperature of 39°C, blood pressure of 124/68 mm Hg, and oxygen saturation in room air of 98% by pulse oximetry. Physical examination revealed pinpoint pupils of bilateral eyes, no corneal reflex, and no doll's eyes signs, but cough and gag reflexes still existed. The deep tendon reflexes were +2 to +3 in the bilateral upper and lower extremities, and the Babinski reflexes were present bilaterally. The Glasgow Coma Scale score was 6(1 for eye open, 1 for verbal, and 4 for motor). Refractory high fever, even under aggressive antipyretic treatment, persisted. The initial laboratory tests revealed a white blood cell count of 12,060/µL, with a differential count of 25% band forms, 67% neutrophils, and 6% lymphocytes. The hemoglobin was 13.1 mg/dL, and the platelet count was 233 X 10[sup3]/µL. Abnormal liver function with aspartate aminotransferase 291 IU/L and alanine aminotransferase 244 IU/L was noted, but there was no hyperammonemia, coagulopathy, or hypoglycemia. The serum MB isoenzyme/total creatine kinase level was markedly elevated to 96/1500 U/L (6%), and C-reactive protein was 4 mg/dL (normal limit < 1 mg/dL). Lumbar puncture revealed the initial pressure of 265 mm H[sub2]O and clear cerebrospinal fluid containing white blood cells 2 cells/µL, red blood cells 1 cell/µL, protein 113 mg/dL, and glucose 118 mg/dL (serum glucose 187 mg/dL). Cranial computed tomography (CT) showed low-density change and the appearance of swelling mainly involving the pons, midbrain, and bilateral thalamus region (Figure 1). Brain magnetic resonance imaging (MRI) revealed high signal intensity on T[sub2]-weighted image mainly in the pons, midbrain and hypothalamus and extension to the bilateral thalamus and posterior limb of the bilateral internal capsule (Figure 2A). Several tiny high signals in the subcortical region of the bilateral high frontoparietal region and the bilateral periventricular white-matter region were also noted.

Intravenous immunoglobulin with a dosage of 1 g/kg/day was administered for 2 days, and refractory high fever subsided 24 hours thereafter. Penicillin and ceftizoxime were administered for 5 days until bacterial culture of the blood, cerebrospinal fluid, and urine showed negative results. Electroencephalography arranged on the ninth day of illness revealed a continuous slow wave at 1 to 2 Hz without any state change. Thrombocytosis with platelets up to 802 X 10[sup3]/µL developed from the 12th day of illness and gradually decreased to the normal range in 1 month. Repeated lumbar puncture on the 14th day of illness revealed 6 white blood cells/µL and 4400 red blood cells/µL on a traumatic lumbar puncture with normal open pressure of 125 mm H[sub2]O and protein levels of 48 mg/dL. Unfortunately, urinary tract infection with acute pyelonephritis occurred, and neurogenic bladder persisted, with difficulty removing the Foley catheter. Direct immunofluorescent testing of the nasopharyngeal aspirate for respiratory viruses was positive for influenza B. and it was later confirmed by cell culture. Complement fixation antibody to influenza B virus was negative (titer < 1:8) at presentation. A convalescent serum sample, obtained 4 weeks after presentation, was positive at a titer of 1:8. Tests for Japanese encephalitis virus, Epstein-Barr virus, and Mycoplasma pneumonia were all negative. Brain MRI done after 1 month of hospitalization revealed much regression of the previous edematous change, with high signal intensity on the T[sub2]-weighted image but high signal intensity change on the T[sub1]-weighted image and slightly high signal intensity change on the T[sub2]-weighted image over the lesion site, in favor of hemorrhagic change (Figure 2B). A hyperintensity lesion over the bilateral central white matter and subcortical region of the bilateral high frontal parietal area showed no obvious interval change. The patient stayed at our hospital for a 2-month period for the rehabilitation programs and complete treatment of the pyelonephritis and was thereafter transferred to the chronic care center for continuing management. On the day of transfer, she seemed more responsive but still suffered severe motor and mental impairment.

Acute neorotizing encephalopathy of childhood was commonly used to describe this novel disease entity after a collaborative report led by Mizuguchi et al in 1995.&sup1,2; Acute necrotizing encephalopathy of childhood is a clinical neuroradiologic entity, and the diagnostic criteria proposed by Mizuguchi et al included the following: (1) acute noninflammatory encephalopathy with an alteration in consciousness levels; (2) multifocal lesions symmetrically distributed in brain regions, including the thalamus, demonstrated by CT or MRI; and (3) the absence of any other reasonable explanation for the cerebral abnormalities.&sup2,6; The clinical onset is usually preceded by viral infection, and the disease affects young children of both sexes, mostly in winter. The laboratory examination sometimes reveals increased serum aspartate transaminase, alanine transaminase, and lactic dehydrogenase without hyperammonemia and hypoglycemia. The concentrations of protein and myelin basic protein in cerebrospinal fluid are increased with or without pleocytosis. The most characteristic features of acute necrotizing encephalopathy of childhood are the neuroradiologic findings with multifocal and symmetric involvement of the thalamus, brain stem, and cerebellum demonstrated by CT, MRI. and even ultrasonography showing homogeneous hyperechogenicity of the panthalamic lesions.&sup2,6; On the basis of the typical clinical manifestations and characteristic neuroimaging findings in our case, a diagnosis of acute necrotizing encephalopathy of childhood can be made.

A recent report from Japan found that at least 20% of patients with influenza encephalopathy exhibited as acute necrotizing encephalopathy of childhood with multifocal, symmetric brain lesions affecting the thalami bilaterally on neuroimaging, and this was recently proposed to be one of the features of influenza-associated encephalopathy.&sup2,6; A number of cases describing the same disease can be classified as acute necrotizing encephalopathy of childhood based on neuroimaging and/or as "influenza" encephalopathy based on virology." However. it is important to identify this kind of influenza-associated encephalopathy presenting with acute necrotizing encephalopathy of childhood because of its high mortality rate and severe neuromuscular sequelae. Influenza encephalopathy can develop by the same pathogenetic mechanisms as acute necrotizing encephalopathy of childhood, possibly via vasoactive substances to local breakdown of the blood-brain barrier or a process leading to vasoconstriction in the central nervous system.[sup7,8] Recently, it has been reported that cytokines such as soluble tumor necrosis factor receptor-1, interleukin-lβ, and interleukin-6 could mediate the disease, and a high plasma concentration of interleukin-6 could be an indicator of progression to encephalopathy.[sup9,10] We checked our patient's interleukin-6 levels at the acute and convalescent, phases, and interleukin-6 was not elevated in these two phases. We presumed that it might be suppressed by our intravenous immunoglobulin therapy, via modulation of the production of cytokines, including interleukin6.[sup11] Arresting the hypercytokinemic process might be effective in preventing the disease progression from causing further damage to the central nervous system.

Approximately 90% of acute necrotizing encephalopathy of childhood cases demonstrate antecedent upper respiratory infection, and the identified or presumed pathogens responsible for these infections have always been viruses.² Influenza A, exanthema subitum, and influenza B are the most common pathogens, although other viral diseases, such as measles, herpes simplex, varicella, and rubella, have also been reported.&sup2,6,7; Because a variety of viruses predispose individuals to acute necrotizing encephalopathy of childhood. it remains unknown whether these viruses have the same pathophysiologic processes or an unidentified agent causes acute necrotizing encephalopathy of childhood. An increased awareness of influenza A and B viruses as causes of acute necrotizing encephalopathy of childhood is necessary for this devastating disorder, and studies to determine the incidence of influenza virus encephalopathy, especially during annual epidemics, should be undertaken. According to the experiences of Wang and Huang in Taiwanese children, 80% of the cases of acute necrotizing encephalopathy of childhood were in children younger than 2 years of age. and the mean age at onset was 20 months.³

The clinical entity of influenza-associated encephalopathy has not gained universal recognition, and many texts on pediatrics do not describe this syndrome. However, influenza-associated encephalopathy has been frequently reported as a complication of influenza in Japanese and Taiwanese children and has become a serious problem during the past several years.12 Acute necrotizing encephalopathy predominantly affects children living in Japan and Taiwan, and a strong possibility of association with genetic or epigenetic factors in these countries is suggested.³ Only a few cases of influenza-associated encephalopathy in children similar to those reported in Japan and Taiwan have been reported from the United States, England, and Australia It could be missed in these countries because most pediatricians do not regard the influenza virus as a causative agent of acute encephalopathy other than Reye's syndrome.[sup12,13] Actually, acute necrotizing encephalopathy of childhood associated with influenza infection reported in Japan and Taiwan is distinct from Reye's syndrome. The clinical differences in patients with influenza-associated encephalopathy include the following: (1) no history of taking aspirin; (2) rapid loss of consciousness, with coma ensuing within 24 hours; (3) in most patients, convulsions shortly after onset; (4) commonly increased cerebrospinal fluid protein; and (5) absence of hyperammonemia and hypoglycemia. Moreover, neuroimaging often reveals bilateral thalamic necrosis as acute necrotizing encephalopathy of childhood.[sup4,7] The clinical manifestation of our patient is compatible with acute necrotizing encephalopathy of childhood according to the diagnostic criteria, instead of Reye's syndrome.

Regrettably, there is still no specific therapy or prevention for acute necrotizing encephalopathy of childhood. In severe cases, intensive supportive care is mandatory, including the management of intracranial and systemic arterial pressure and seizure control.² Although amantadine is available to treat influenza A infections, and oseltamivir was recently licensed for treatment of influenza A and B infections, there is still very limited information in patients with acute necrotizing encephalopathy of childhood-associated influenza virus infections.[sup7] It is reasonable that administration of corticosteroid could decrease vascular permeability at the acute stage, and intravenous immunoglobulin might modulate the production of cytokines, including interleukin-1, interleukin-6, and tumor necrosis factors, to arrest the hypercytokinemic process of acute necrotizing encephalopathy of childhood. However, their clinical usefulness has not yet been determined systematically and needs further evaluation.&sup2,11; The administration of intravenous immunoglobulin seemed to work on our patient in controlling the refractory high fever and in stabilizing the disease progression.

The prognosis is often grave, and the mortality rate ranges from 25 to 30%.&sup2,11; Those who survive acute necrotizing encephalopathy of childhood often suffer from mild to moderate motor and mental sequelae, and less than 10% of patients recover completely, such a recovery being more likely in children older than 4 years without elevated aminotransferases and brainstem lesions on neuroimaging.&sup2,3,14; Thrombocytopenia and elevated serum levels of aspartate transaminase, alanine transaminase, lactic dehydrogenase, and creatine phosphokinase levels, as well as prolonged coagulation times, have recently been thought to be factors associated with a poor prognosis.[sup15] These features are compatible with the condition of our presenting case. Typical acute necrotizing encephalopathy of childhood usually shows irreversible multifocal involvement, including the thalamus, and leaves severe sequelae in spite of treatment, but some patients show a good prognosis with reversible involvement, and complete recovery, which are evidenced by the time course of neuroimaging findings, despite the term "necrotizing."&sup2,6; The reasons for the various outcomes of acute necrotizing encephalopathy of childhood remain unknown, as does the etiology itself. Yoshikawa et al recently found that a mild form of acute necrotizing encephalopathy of childhood could have the character of transient unilateral thalamic involvement.[sup6]

Many important issues remain to be clarified regarding the pathogenesis and treatment of acute necrotizing encephalopathy of childhood associated with influenza infections, especially considering its prevalence in East Asia, high mortality, and severe neuromuscular sequelae in survivors. Our patient is very young, and such children with their first influenza infection could be more at risk of influenza-associated encephalopathy presenting with acute necrotizing encephalopathy of childhood.[sup16] Theoretically, vaccination of children less than 5 years of age prior to the epidemic season will prevent influenza-associated encephalopathy or encephalitis.

Received April 28, 2003. Accepted for publication July 15, 2003.

PHOTO (BLACK & WHITE): Figure 1. Brain computed tomographic scan shows low-density change and swelling involving bilaterally the thalamus region.

PHOTO (BLACK & WHITE): Figure 2. A, Brain magnetic resonance imaging (MRI) revealed low signal intensity on T.- (a) and high signal intensity on T[sub2]- (b) weighted axial images in the bilateral thalami and posterior limb of the bilateral internal capsule. A, T[sub1]-weighted coronal image (c) revealed high signal intensity in the pons, midbrain, hypothalamus, and bilateral periventricular white-matter region. B, The brain MRI after 1 month of hospitalization revealed high signal intensity change on T[sub1]-weighted axial (a) and coronal (c) images over the pons, midbrain, and bilateral thalami, in favor of hemorrhagic change, and slightly high signal intensity on a T[sub2]-weighted axial image (b).

1 Mizuguchi M, Abe J, Mikkaichi K, et al: Acute necrotizing encephalopathy of childhood: A new syndrome presenting with multifocal, symmetrical brain lesions. J Neural Neurosurg Psychiatry 1995;58:555-561.

2. Mizuguchi M: Acute necrotizing encephalopathy of childhood: A novel form of acute encephalopathy prevalent in Japan and Taiwan. Brain Dev 1997; 19:81-92.

3. Wang HS, Huang SC: Acute necrotizing encephalopathy of childhood. Chang Gung Med J 2001 ;24:1 10.

4. Shinjoh M, Bamba M, Jozaki K, et al: Influenza type associated encephalopathy with bilateral thalamic necrosis in Japan. Clin Infect Dis 2000;31:611-613.

5. Yoko J, Motohiro S, Masumi T, et al: Influenza A virus encephalopathy with symmetrical thalamic lesions. J Pediatr 2000;159:319-321.

6. Yoshikawa H, Watanabe T, Tokinari A, Oda Y: Clinical diversity in acute necrotizing encephalopathy. J Child Neural 1999; 14:249-255.

7. Sugaya N: Influenza-associated encephalopathy in Japan: Pathogenesis and treatment. Pediatr Int 2000;42:215-218.

8. Shinjoh M, Sugaya N, Takahashi K, et al: A case of influenza A virus associated encephalopathy with bilateral thalamic hemorrhage. Kansenshogaku Zasshi 1999;73:778-781.

9. Togashi T, Matsuzono Y, Itakura O, et al: IL-6 and TNF-alpha in cerebrospinal fluid from infantile encephalitis-encephalopathy patients during influenza seasons. Nippon Shonika Gakkai Zasshi 1999; 103:16-19.

10. Ito Y, Ichiyama T, Kimura II, et al: Detection of influenza virus RNA by reverse transcription-PCR and proinflammatory cytokines in influenza-virus-associated encephalopathy. J Med Viral, 1999; 58:420-425.

11. Kazatchkine MD, Kaveri SV: Immunomodulation of autoimmune and inflammatory diseases with intravenous immune globulin. N Engl J Med 2001:345:747-755.

12. Sugaya N, Yoshikawa T, Miura M, et al: Influenza encephalopathy associated with infection with human herpesvirus 6 and/or human herpesvirus 7. Clin Infect Dis 2002;34:461-166.

13. Sugaya N, Miura M: Amantadine therapy for influenza type A associated encephalopathy. Pediatr Infect Dis J 1999;18:734.

14. George KS: Complications of influenza infection in children. Pediatr Ann 2000; 11:683-690.

15. Morishima T, Togashi T, Yokota S, et al: Encephalitis and encephalopathy associated with an influenza epidemic in Japan. Clin Infect Dis 2002;35:512-517.

16. Fujimoto S, Kobayashi M, Uemura O, et al: PCR on cerebrospinal fluid to show influenza-associated acute encephalopathy or encephalitis. Lancet 1998;352:873-875.