ELECTRONIC ARTICLE |


,||
* Section of Neurology
Section of Infectious Disease
|| Diagnostic Virology Laboratory, Department of Pediatrics, Texas Childrens Hospital, Houston, Texas
¶ Department of Neurology, Division of Neuroscience, Baylor College of Medicine, Houston, Texas
Influenza Branch, Division of Viral and Rickettsial Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Atlanta, Georgia
| ABSTRACT |
|---|
|
|
|---|
Methods. We reviewed the medical and laboratory records of all children who were hospitalized with neurologic symptoms and who also had evidence of influenza virus infection by rapid antigen testing or viral isolation.
Results. Eight children aged 5 months to 9 years with neurologic complications associated with influenza A were identified. None of the children had received the influenza vaccine. Four presented with seizures, 3 with mental status changes, and 1 with mutism. All but 1 of the patients had influenza A viral antigen detected in nasal wash samples. Influenza A virus was isolated in culture from nasal wash specimens obtained from 6 of the patients; influenza A virus was also isolated from the cerebrospinal fluid of 1 of these patients. None of the patients had serum metabolic abnormalities or other cerebrospinal fluid abnormalities. Three of the patients had brain imaging abnormalities. Five of the patients were treated with antivirals. All 8 of the patients survived, 6 with complete recovery and 2 with sequelae (1 mild and 1 severe).
Conclusions. Neurologic symptoms and sequelae were associated with influenza A virus infection in children during the 20032004 influenza season in Houston, Texas. Influenza should be considered in the differential diagnosis in patients with seizures and mental status changes, especially if they present with respiratory symptoms or during an influenza outbreak.
Key Words: encephalopathy altered mental status seizure viral infection infectious complication
Abbreviations: TCH, Texas Children's Hospital RT-PCR, reverse transcriptase-polymerase chain reaction CSF, cerebrospinal fluid EEG, electroencephalogram MRI, magnetic resonance imaging CT, computed tomography RSV, respiratory syncytial virus HSV, herpes simplex virus ANE, acute necrotizing encephalopathy CNS, central nervous system
Influenza A and B viruses cause influenza epidemics each fall and winter in the United States. Infection typically causes an acute febrile respiratory illness (nonproductive cough, sore throat, nasal congestion, and rhinorrhea) accompanied by headache and myalgias. Children may also experience gastrointestinal disturbances (diarrhea, vomiting, and anorexia).1 Lower respiratory syndromes such as bronchiolitis, croup, bronchitis, and pneumonia may be severe, and, in rare cases, influenza can progress to life-threatening respiratory compromise, either from the primary infection itself or because of a complicating bacterial superinfection. Other serious complications of influenza include myositis, rhabdomyolysis, and myocarditis. An average of >36 000 influenza-attributable deaths occur during seasonal epidemics in the United States each year.2 In addition, influenza pandemics have been associated with far greater morbidity and mortality than annual epidemics. For example, the 1918 influenza pandemic is estimated to have caused >20 million deaths worldwide, including >500 000 deaths in the United States.3 During the 20032004 influenza season in the United States, 142 deaths occurred among children.4
Neurologic complications associated with influenza have been reported for >100 years. Most recently, epidemic influenza in Japan in the late 1990s was associated with hundreds of cases of encephalopathy among children.5,6 Although the majority of the patients had relatively minor symptoms, a small but significant number experienced serious complications, resulting in neurologic sequelae and even death. Small numbers of children with neurologic complications have been reported sporadically in Europe and Canada, whereas reports from the United States in the past 60 years have been exceedingly rare.710
During mid-October to early November 2003, an uncharacteristically early and severe outbreak of influenza A occurred in Houston, Texas. An unusually high number of laboratory-confirmed influenza A cases were medically evaluated and patients were hospitalized at Texas Children's Hospital (TCH), including some cases associated with a variety of neurologic symptoms. We report here the largest single-season case series of influenza-associated neurologic complications from an American pediatric center.
| METHODS |
|---|
|
|
|---|
1 of the authors during the recent early autumn influenza outbreak were included. Clinical data from medical charts and neuroimaging studies were reviewed. All neurologic examinations were performed by board-certified pediatric neurologists (G.C. and T.E.L.) and/or fellows in the Accreditation Council for Graduate Medical Education -accredited pediatric neurology training program (J.N. and S.M.M.).
Virology
Rapid antigen testing for influenza A and B viruses was performed by the Diagnostic Virology Laboratory at TCH, using a commercially available lateral flow immunoassay11 (NOW Flu A/B; Binax, Inc, Portland, ME). The test was performed on nasal wash specimens obtained by a standardized protocol using the manufacturer's instructions, with internal kit positive, negative, and procedural controls, as well as external, in-house laboratory controls for each test kit. Results of these rapid tests were reported to clinicians within 2 hours. All specimens submitted for rapid testing were also inoculated for virus isolation onto human foreskin fibroblast, Rhesus monkey kidney, and human lung carcinoma (A549) cell culture monolayers. Viral cultures were inspected daily under light microscopy for cytopathic effect. Hemadsorption with a 0.4% suspension of guinea pig red blood cells was performed on days 2, 5, and 14 of incubation of Rhesus monkey kidney cell cultures. Virus identification and typing of hemadsorption-positive cultures was confirmed by immunofluorescence assay (Chemicon International, Temecula, CA). Influenza A virus isolates were subtyped and antigenically characterized further by the Strain Surveillance Section, Influenza Branch, of the Centers for Disease Control and Prevention (Atlanta, GA), either by hemagglutination inhibition using postinfection ferret antisera to determine antigenic relatedness to known reference strains of influenza A virus or by reverse transcriptase-polymerase chain reaction (RT-PCR) assay.
| RESULTS |
|---|
|
|
|---|
|
Patient 2
A previously healthy 9-year-old black boy with attention-deficit/hyperactivity disorder and otherwise normal development complained of subjective fever, frontal headache, and fatigue on the day before admission. On the morning of admission, he also complained of neck pain and photophobia. Shortly after arising, he experienced a generalized tonic-clonic seizure that lasted for 1.5 minutes, followed by a postictal period first of sleep (for 20 minutes) then of decreased responsiveness and continued photophobia that lasted for 7 to 8 hours. His mental state then returned to normal. His home medications were methylphenidate and dexmethylphenidate, and no recent dosing changes had been made. During emergency medical services transport to TCH, the patient was noted to have a fever of 102°F. On admission, he remained febrile; general examination was remarkable only for a 12/6 systolic ejection murmur. On neurologic examination, the patient was alert and oriented and had mild neck stiffness, a positive Brudzinski sign, a negative Kernig sign, and some minor difficulty with tandem gait. A computed tomography (CT) scan of the brain showed decreased size of the sulci, sylvian fissures, and ventricles, and MRI additionally revealed a mild gyral swelling in the left frontal region; both studies were believed to be consistent with a diagnosis of meningoencephalitis. Vancomycin and cefotaxime were administered intravenously on admission. Urine drug screen, CSF viral and bacterial cultures, and respiratory syncytial virus (RSV) rapid antigen test on nasal wash were negative. Rapid antigen test and viral culture from nasal wash were positive for influenza A virus. Oseltamivir was started on day 2 of hospitalization and was administered for 5 days. His mental status remained at his normal baseline, but he continued to have a minor headache and increased duration of sleep cycles.
Patient 3
A 6-year-old right-handed black girl with a history of ventricular septal defect (surgical closure at 3 years of age), mild congestive heart failure, and asthma presented with a 2- to 3-week history of intermittent rhinorrhea and cough. On the day of admission, she had a fever to 105°F accompanied by a fluctuating mental status that at times consisted of staring, moving her eyes aimlessly, inability to recognize family members, and incoherence. No seizure activity was noted. In the emergency department, she had 4 episodes of bilious, nonbloody emesis as well as 1 episode of diarrhea, followed by an acute oxygen desaturation into the 80s that responded well to O2 by nasal canula. There were no complaints of stiff neck, headaches, diplopia, blurred vision, difficulty swallowing, or shortness of breath. The patient's home medications were albuterol and inhaled budesonide for asthma. Temperature on admission was 103.6°F; O2 saturation was 100% on 10 L of oxygen and 84% on room air. General physical examination revealed intermittent rigors, tonsillar erythema without exudates, and diffuse crackles on lung examination. On mental status testing, the patient was lethargic and irritable but was intermittently able to follow 1-step commands and to recognize close family members. The remainder of the neurologic examination was within normal limits. Blood and CSF tests were normal; urine drug screen, RSV antigen test of nasal wash, and specific CSF studies (herpes simplex virus [HSV] PCR, West Nile, and mycoplasma cultures) were negative. Viral culture of a nasal wash specimen yielded influenza A virus. Chest x-ray was consistent with obstructive airway disease versus aspiration. A CT scan of the head demonstrated sulci, ventricles, and cisterns that were at the lower size limit for age. Nasal wash was positive for influenza A by rapid antigen test. The patient's mental status returned to baseline within 24 hours of admission. The patient received 1 dose of vancomycin, and cefuroxime and acyclovir were administered. Rimantadine was administered on day 2 of hospitalization, and acyclovir was discontinued. The patient completed a 5-day course of rimantadine and was discharged from the hospital 7 days later without neurologic sequelae.
Patient 4
A 14-month-old white girl, previously healthy except for a history of eczema, had a 2-day history of cough, clear rhinorrhea, and fever to 102°F at home. On the morning of admission, she was increasingly lethargic and had poor oral intake. On the way to her pediatrician's office, she had 2 generalized tonic-clonic seizures that lasted 1.5 minutes each, followed by postictal sleepiness. She was noted to be febrile at the physician's office, and ibuprofen was administered. On transport to TCH, she had several short generalized seizures, which she continued to have every 20 to 30 minutes (several of which were not associated with fever) after arrival in our emergency department. The patient was loaded with lorazepam, fosphenytoin, and phenobarbital with cessation of clinical seizure activity. Vital signs on admission were notable for a temperature of 103°F. General examination was remarkable for a maculopapular rash over the trunk and proximal extremities. Neurologic examination demonstrated a lethargic infant with waxing and waning mental status and truncal ataxia. Routine labs and CSF studies were unremarkable. RSV antigen test, CSF viral and bacterial cultures, and enterovirus and HSV PCR of CSF were negative. Nasal wash was positive for influenza A by rapid antigen test, and influenza A virus was isolated from nasal wash culture. CT and MRI scans of the brain were within normal limits. EEG shortly after admission showed multiple electrographic seizure discharges arising from the left central region. Two repeat EEGs over the next 24 hours demonstrated bilateral central spike foci but no seizure activity, and fosphenytoin and phenobarbital were discontinued after 48 hours. However, on the third day of admission, the patient began having multiple tonic-clonic seizures that lasted <30 seconds and did not respond to additional phenobarbital administration. Fever was associated with some but not all of the seizure activity. On day 6 of admission, oxcarbazepine was started and seizure activity stopped. The patient completed a 5-day course of rimantadine and was discharged from the hospital on oxcarbazepine 10 days after admission. She was at her neurologic baseline at the time of discharge.
Patient 5
A 5-month-old white girl with a history of gut malrotation and successful surgical treatment presented with a 2-day history of vomiting and upper respiratory infection. On the day of admission, she was noted to be excessively sleepy and presented to the TCH emergency department. She was believed to be slightly dehydrated but remained lethargic despite adequate rehydration. Over the next 12 hours, the patient did not respond to voice, sternal rub, or painful procedures such as intravenous catheter placement. She then became more responsive and opened her eyes to voice. The neurologic examination was otherwise normal, and her development to date has been normal. Influenza A rapid antigen test on nasal wash was positive. The patient was discharged 2 days after admission at her neurologic baseline.
Patient 6
A 5-year-old right-handed Hispanic boy presented with new onset of seizures. One day before admission, he was noted to have fever, abdominal cramps, and vomiting. On the day of admission, he had a nonfebrile, left focal motor seizure that secondarily generalized with tonic-clonic activity for a total duration of 2 minutes. After recovery from a postictal confused state, his examination was nonfocal. He was empirically started on cefotaxime and acyclovir. Initial EEG demonstrated diffuse slow activity with no occipital dominant rhythm. A noncontrast brain CT on the day of admission was normal. An enhanced MRI scan of the brain obtained 24 hours after the seizure showed bilateral increased T2-weighted and fluid attenuated inversion recovery signal and mild swelling in the hippocampal heads and bodies. In addition, increased T2 signal with restricted diffusion was noted in the splenium and genu of the corpus callosum (Fig 1). These imaging findings were interpreted as likely secondary to the seizure activity, although acute disseminated encephalomyelitis was also noted as a possibility. Rapid antigen testing for influenza was negative, but subsequent culture from a nasal wash specimen identified influenza A. Mycoplasma, Epstein-Barr virus, and HSV PCR testing were negative. Serum electrolytes, glucose, renal function studies, and complete blood count were normal. Routine blood, urine, and CSF cultures were negative. Lumbar puncture was traumatic with red cell count of 196; white cell count was 6 with normal protein and glucose. Additional investigations included very-long-chain fatty acid testing, which was normal. The patient had no additional seizures and did not receive anticonvulsants. He was discharged from the hospital with a normal neurologic examination.
|
Two days into the hospital course, the patient began to show some improvement in his speech, with production of single words, although with hypernasal quality, slowed rate, and low volume. He was able to eat small amounts of pureed food without difficulty. On hospital day 3, he had some mild choreiform movements of the trunk and extremities that were not debilitating. He was discharged from the hospital on hospital day 4. At the time of follow-up evaluation 3 weeks after discharge, his speech was improved, although still slow and low in volume. The remainder of his examination demonstrated the baseline nystagmus and some persistent mild choreiform movements of the fingers on horizontal suspension. Repeat EEG done 3 months after presentation was unchanged.
Patient 8
A 13-month-old Hispanic girl with normal development and a history of 1 previous febrile seizure presented with a 2-day history of fever and generalized status epilepticus in July 2003 that required brief intubation. A head CT scan demonstrated decreased attenuation in the left caudate and bilateral thalami. MRI scan showed extensive increased T2 signal bilaterally in the thalami, caudate nuclei, deep cerebellar nuclei, corpus callosum (genu and splenium), left putamen, hippocampus, and cerebral peduncle (Fig 2 A and B). The lesions did not enhance after administration of contrast. CSF studies and lactate were normal. Mycoplasma titers, routine blood and urine cultures, CSF cultures, and HSV and enterovirus PCR were negative. Serum electrolytes, lactate, and hematocrit were also normal. White blood cell count was low (1.69), and alanine transaminase was slightly elevated (138). Urine amino and organic acids, serum amino acids, free/total carnitine ratio, and mitochondrial DNA analysis were normal. The child was empirically treated with acyclovir, vancomycin, and cefotaxime. Intravenous dexamethasone (4 mg/kg per day for 5 days) was started for presumptive acute disseminated encephalomyelitis. The child improved and was discharged from the hospital at her neurologic baseline. She continued to acquire new skills after discharge.
|
| DISCUSSION |
|---|
|
|
|---|
All of the neurologic presentations in our cases were associated with influenza A virus infection and were, for the most part, in the milder end of the encephalopathic spectrum. None of the children were Asian in descent. All occurred within 3 weeks of upper respiratory infection symptom onset, with 6 of 8 occurring within the first week. All of the patients also had clinical symptoms and signs of influenza. The timing and concurrent influenzal symptoms suggest that acute influenza caused by the influenza A virus, rather than a "postinfluenza encephalopathy,"14,16 was responsible for the associated neurologic findings.
Seizures were the most common neurologic symptom in our patients and were seen in 4 of 8 patients. However, we do not believe that the seizures seen in these cases can be explained simply as febrile seizures. Three of our patients (1, 2, and 6) had laboratory and/or imaging evidence of central nervous system (CNS) involvement. Influenza A virus was isolated from the CSF of patient 1; patient 2 is 9 years old, and patient 6 is 5 years old, making a first episode of febrile seizures unlikely; and patients 2 and 6 had brain imaging studies suggestive of direct CNS involvement. Finally, patient 4 experienced recurrent seizures 3 days after the first seizure, and several of these were in the absence of fever. Although we do not believe fever to be the cause of seizures in these children, it has been reported that influenza virus infection is more commonly associated with febrile seizures in children than are other viral infections.13 The mechanism by which this occurs is currently unknown.
Patient 8's neurologic problems could have been attributable to a metabolic or mitochondrial disorder that was exacerbated by influenza; the similar previous episode that was associated with multiple brain lesions on MRI supports this contention. However, extensive evaluations for other infectious and metabolic causes, including a muscle biopsy, were negative, and the second episode was clearly associated with influenza A virus infection. We believe that this patient's second presentation is best explained as influenza-associated ANE, representing only the second such report from the United States.30 The only other associated infectious finding was the isolation of adenovirus in the child's stool at the second presentation. It is interesting that Steininger et al14 reported a possible association between adenoviral infection and postinfluenzal encephalopathy after influenza A virus infection. Our patient's first episode was also associated with a possible viral prodrome. There are reports of children having multiple episodes of influenza-related ANE.28 Whether these represent true viral syndromes or a currently undetectable inborn metabolic derangement manifesting under stress is an open question. Regardless, it is important to recognize that influenza virus infection can trigger such a response, even in the absence of CNS infection.
It has been postulated that influenza encephalopathy might be related to metabolic disturbances that occur in relation to the primary infection. In support of this argument, influenza, particularly caused by influenza B virus infection, has been associated with the occurrence of Reye's syndrome in children exposed to aspirin.31 None of our patients had a history of aspirin administration, and, in the cases in which liver enzymes were analyzed, none had evidence of hepatic failure. This, coupled with the clinical presentations and subsequent improvement in the majority of cases, makes it highly unlikely that any of these patients had Reye's syndrome.
The pathogenic mechanism of influenza virus infection on the nervous system remains a topic of debate. In fact, very few studies have shown the presence of viral antigens in CNS tissue at autopsy, and isolation of virus from CSF has been equally rare. All 8 of the cases in this series had influenza A virus detected in respiratory specimens. However, we were able to culture influenza A virus from the CSF of only 1 patient (patient 1). Recently, RT-PCR techniques have been used to test for the presence of viral RNA in serum and CSF samples; however, these techniques have met with varied success, and the majority of studies have been unable to provide direct evidence of influenza virus infection in CNS sites.10,14 RT-PCR for influenza A viral RNA was not done on any of the CSF samples in this case series because priority was given to viral culture techniques, leaving insufficient quantities for PCR.
Influenza A virus strains with certain hemagglutinin (H) and neuraminidase (N) profiles are more often associated with CNS complications. In particular, influenza A (H3N2) virus infections have been found in the majority of encephalopathy cases reported in Japan and Europe, although occasionally influenza A (H1N1) viruses have been isolated as well. All of the viral isolates (from patients 14, 6, and 8) in this series were antigenically characterized as A/Fujian/411/2002(H3N2)-like. This strain is an antigenic drift variant of the influenza A (H3N2) vaccine strain, A/Panama/2007/99 (H3N2)-like. Whether the association between certain influenza A virus strains and neurologic complications results from an undetected tropism of these strains for the CNS or a greater likelihood of causing a host reaction that leads to encephalopathy remains unknown.
It is unknown whether influenza vaccination would have prevented or attenuated the neurologic complications associated with influenza A virus infection in the reported cases. Four of the 8 children were in a high-risk group currently recommended for influenza vaccination.32 However, none of the patients was vaccinated, and most of them presented before the influenza vaccine was widely available in the Houston area. In addition, the 20032004 influenza vaccine did not include the A/Fujian/411/2002 (H3N2)-like strain, which was the predominant strain associated with the epidemic and identified in 6 of the cases.
Of equal ambiguity is the utility of antiviral treatments on the course or outcome of influenza-associated encephalopathy or encephalitis. Efficacy of antivirals for reducing the symptoms, signs, and duration of acute uncomplicated influenza (respiratory disease) has been established. No randomized, controlled study has examined the impact of antiviral treatment on influenza-associated CNS manifestations, although the use of antivirals in some neurologic complications of influenza has been reported.5,10,33 Several of the patients in our series were treated with a full course of antiviral medication (rimantadine or oseltamivir). It is not clear whether this intervention resulted in any clinical improvement or whether the neurologic symptoms were self-limited. It seems reasonable to recommend antiviral therapy at this time, but additional studies are needed to evaluate its true role in the treatment of influenza-related encephalopathy.
In summary, we report 8 influenza A cases that were associated with encephalopathy syndromes during a severe influenza outbreak in Houston, Texas. Our experience with influenza during the fall of 2003 was similar to that reported in previous years in Japan but distinctly different from our previous experience. Whether this reflects the virulence of the predominant influenza strain A/Fujian/411/2002 (H3N2)-like or whether high attack rates resulted in the identification of a number of uncommon neurologic complications is unknown. Influenza-associated encephalopathy should be considered in the differential diagnosis for children who present with new-onset seizures or mental status changes during the influenza season.
| FOOTNOTES |
|---|
Reprint requests to (S.M.M.) Fellow in Child Neurology, CCC 17.10, 6621 Fannin St, Houston, TX 77030. E-mail: maricich{at}bcm.tmc.edu
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. D. de Jong, B. V. Cam, P. T. Qui, V. M. Hien, T. T. Thanh, N. B. Hue, M. Beld, L. T. Phuong, T. H. Khanh, N. V. V. Chau, et al. Fatal Avian Influenza A (H5N1) in a Child Presenting with Diarrhea Followed by Coma N. Engl. J. Med., February 17, 2005; 352(7): 686 - 691. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||