In April 2009, a novel influenza A (H1N1) virus was identified in Mexico and the United States.1 Within a few weeks, worldwide spread of the virus led the World Health Organization to declare a pandemic. Newly diagnosed infections decreased in the northern hemisphere during summer, but a second wave of new cases swept through the United States, peaking in most parts of the country by mid-October. Seasonal influenza is known to cause neurological complications in children, including seizures, encephalopathy, myositis, and acute necrotizing encephalitis. Although small case series have described a variety of neurological complications in association with 2009 H1N1 influenza,2–4 it is unclear if the incidence and severity of neurologic complications differs from seasonal influenza. The purpose of our study was to determine the incidence, characteristics, and outcomes for neurologic complications related to 2009 H1N1 influenza infection among pediatric patients, compared to seasonal influenza.
The 2009 pandemic influenza A (H1N1) has been recognized to cause neurological complications including seizures and encephalopathy. We identified 18 children with 2009 H1N1 influenza and neurological complications from first and second wave activity, and compared characteristics to seasonal influenza. Seizures, encephalopathy, and status epilepticus were common presentations. Focal neurological symptoms persisted in 22% of patients at discharge. Compared to seasonal influenza, patients with pandemic 2009 influenza were more likely to have encephalopathy, focal neurological findings, aphasia, and abnormal electroencephalographic findings. In addition, we noted a trend toward heightened neurological complications following second wave influenza activity. ANN NEUROL 2010
Patients and Methods
Approval for this study was obtained from the institutional review boards of the University of Utah and Primary Children's Medical Center. We conducted a retrospective study of children hospitalized with 2009 H1N1 influenza infection and neurological complications between April 1, 2009 and November 30, 2009. Patients were included if they were <19 years old at the time of presentation, were hospitalized with confirmed pandemic 2009 H1N1 influenza virus, and had a neurological complication that led to or complicated the hospitalization (and was not secondary to some systemic problem causing a neurological complication). Neurological complications included seizures, febrile seizures, status epilepticus, encephalopathy, encephalitis, myositis, myalgia, aphasia, ataxia, neuropathy, Gullain-Barre syndrome, or other focal neurological complaints. We used children hospitalized with seasonal influenza and neurological complications as a comparison group, for the time period July 1, 2004 through June 30, 2008. The study was conducted at a children's hospital that provides both primary and tertiary care services. The cachement area has an estimated pediatric population of >1 million children distributed among Utah, Idaho, Wyoming, Nevada, and Montana.5 First and second wave activity was defined as April 1 to July 31 and August 1 to November 30, 2009. These ranges for each pandemic wave were based on the incidences of new H1N1 influenza admissions at our institution. We searched records of children known to the neurology service directly, as well as using a computer-based screen of all patients identified with laboratory-confirmed influenza A together with at least 1 of the International Classification of Diseases, Ninth Revision, Clinical Modification codes for a neurological diagnosis (see Supporting Information Methods). The positive predictive value of direct fluorescent antibody (DFA) staining for influenza ranged from 85 to 100%6 and is monitored annually.
From a total of 303 children diagnosed with 2009 H1N1 influenza virus hospitalized between April 1, 2009 and November 30, 2009, 18 met inclusion criteria, with 9 patients each in wave 1 and wave 2 ( Fig). Demographic characteristics were similar in the 2 waves (Table 1). We noted that most patients had an underlying medical condition (83%), primarily neurologic conditions (66%). The mean age was 6.5 years (range, 0.3–15 years). Fifteen (83%) were treated with oseltamivir. All patients had 2009 H1N1 influenza detected by polymerase chain reaction (PCR) from upper respiratory secretions. Cerebrospinal fluid (CSF) was examined for 2009 H1N1 influenza virus by PCR in 1 patient; virus was not detected. Lumbar puncture was performed in 10 patients, and none had pleocytosis or elevated protein.
In comparison, 234 children were hospitalized with seasonal influenza in the 2004 to 2008 influenza seasons. During this time period, any hospitalized patient with respiratory symptoms had DFA testing on upper airway secretions; in addition, patients with unexplained fever also often had DFA testing performed.7 Sixteen patients had neurological complications. Patients with seasonal influenza and neurological complications were younger (mean age, 2.4 years) and had fewer underlying medical conditions (25%). None of the patients with seasonal influenza and neurological complications had encephalopathy, aphasia, or focal neurological deficits.
The median length of hospitalization for patients with neurological complications with both seasonal and 2009 H1N1 influenza was 2 days (Table 2). However, for 2009 H1N1 patients, 10 required intensive care unit admission (4 because of status epilepticus, 6 because of respiratory distress), and 5 patients required mechanical ventilation (2 cases of which were secondary to intubation during status epilepticus). Although not statistically significant, this was more than the seasonal influenza patients (4 patients in intensive care, 3 intubated). One patient with 2009 H1N1 influenza died following respiratory failure and cardiac arrest (no patients with seasonal influenza and neurological complications died).
For 2009 H1N1 influenza, the most common neurologic manifestations were seizures (67%) and encephalopathy (50%) (see Table 2). Among 12 children who experienced seizures, 7 presented in status epilepticus. Electroencephalographic (EEG) abnormalities were seen in 8 of 11 cases, including diffuse slowing, spike-and-wave discharges, or a burst-suppresssion pattern (which were new findings for these patients). Ten children were treated with either novel (to the patient) or additional (in addition to their baseline) antiepileptic medications during their hospitalization. Two patients were discharged home on antiepileptic medications. Magnetic resonance imaging (MRI) of the brain was obtained on 7 patients and was abnormal in 3. In 2 patients, multifocal subcortical white matter (T2 sequence) attenuation was observed. One patient had multifocal T2 signal intensity abnormalities in both cortical and subcortical white matter. No MRIs showed diffusion restriction or contrast enhancement. Due to persistent encephalopathy, 2 children were treated (1 with intravenous steroids, 1 with intravenous immunoglobulin) without improvement. In contrast, for seasonal influenza, brain MRI was normal, and only 1 EEG was abnormal.
Compared to seasonal influenza, patients with 2009 H1N1 influenza were significantly more likely to present with encephalopathy, aphasia, or focal neurological findings. In addition, they were more likely to have abnormal EEG findings. A trend toward increased severity of neurological complications was noted in wave 2 as compared to wave 1 patients, although these did not reach statistical significance (see Table 2).
Pandemic 2009 H1N1 influenza is associated with heightened neurological complications in children compared to seasonal influenza. We observed neurological complications in 18 of 303 hospitalized patients with 2009 H1N1 influenza. Seizures and/or encephalopathy were the most commonly observed neurological complications, and most of the patients had a pre-existing underlying medical (usually neurological) condition. We also found that more patients had neurological deficits and/or required continued treatment with antiepileptic medications at the time of hospital discharge. Of the 4 patients with persistent neurological deficits, 3 had abnormal brain MRI. Two of 3 patients (with persistent neurological symptoms) for whom follow-up was available had complete neurological recovery over the next 6 months. Finally, we noted a nonsignificant trend toward increased neurological complications in the second pandemic wave. Patients with neurological complications from the second wave were more likely to have MRI abnormalities or epileptiform EEGs, to require intensive care unit admission, and to have persistent neurological abnormalities at discharge.
Seasonal influenza is known to cause neurological complications in children at a low rate (from 1 to 4 per 100,000 person-years).8–10 In addition, underlying neurological disease is a known risk factor for influenza infection, both for seasonal influenza and for 2009 H1N1 pandemic influenza.11–13 The etiology of neurological complications in influenza infection is unclear, and some authors suggest an indirect autoimmune reactivity.14 Our data showing an absence of CSF pleocytosis (as well as an inability to detect viral DNA by PCR in the CSF of 1 patient) are consistent with this potential mechanism. We did not observe any benefit from the use of steroids or intravenous immunoglobulin for the treatment of encephalopathy. The absence of proven treatments for influenza-related neurological complications underlines the importance of vaccination, although neurological complications can result from vaccination as well. Although we did not observe any neurological complications from oseltamivir use in our cohort, both we and others have noted the potential for neurological side effects.15
Our study is the most extensive evaluation of neurological complications following pandemic 2009 H1N1 influenza in children, includes a larger sample size than the combined previous case reports,2–4 and reports data from the 2 pandemic waves. Because our institution is the only children's hospital in a multistate region, and is the only provider of in-patient pediatric neurology care and consultation, most cases of pediatric 2009 H1N1 influenza with neurological complications in our region would have been cared for at our institution. Nevertheless, our sample size was too small to demonstrate statistical significance for many of the trends observed: for example, the finding that in the second wave there were 3 premature infants as compared to 1 in the first wave. Direct comparison of neurological complications from seasonal influenza to those from 2009 H1N1 influenza is limited in part by a potential ascertainment bias in that the DFA testing used for seasonal influenza has a lower sensitivity compared to PCR, and may not have been performed on all patients with neurological presentations.
This study demonstrates the heightened neurological complications with pandemic 2009 H1N1 influenza in children. In addition, our observation of possible increased severity of neurological complications during the second pandemic wave should lead to close monitoring of potential neurological morbidities in future pandemic H1N1 influenza.
This study was supported by Centers for Disease Control and Prevention grant 1U18IP000303 (K.A.) and the Primary Children's Medical Center Foundation/Children's Health Research Center, University of Utah (J.L.B. and J.J.E).
We thank A. Pavia for critical reading and helpful comments.
Potential Conflicts of Interest
Nothing to report.