 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ARTICLE |
a Infectious Diseases Division, Montreal Children's Hospital, McGill University Health Centre, Montreal, Quebec, Canada
b Division of Infectious Diseases, Stollery Children's Hospital, Department of Pediatrics, University of Alberta, Edmonton, Alberta, Canada
c Division of Infectious Diseases, British Columbia Children's Hospital, Vaccine Evaluation Center, University of British Columbia, Vancouver, British Columbia
d Department of Pediatrics, Dalhousie University, IWK Health Centre, Halifax, Nova Scotia, Canada
e Centre Hospitalier Universitaire de Québec, Laval University, Québec, Québec, Canada
f Division of Infectious Diseases, Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada
g Winnipeg Children's Hospital, University of Manitoba, Winnipeg, Manitoba, Canada
h Department of Pediatrics, Hôpital Ste Justine, University of Montréal, Montréal, Québec, Canada
i Division of Infectious Diseases, Children's Hospital of Eastern Ontario, University of Ottawa, Ottawa, Ontario, Canada
j Provincial Health Office, Alberta Health and Wellness, Edmonton, Alberta, Canada
k Immunization and Respiratory Infections Division, Centre for Infectious Disease Prevention and Control, Public Health Agency of Canada, Ottawa, Ontario, Canada
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
METHODS. All of the children hospitalized with laboratory-confirmed influenza at 9 Canadian tertiary care hospitals during the 2003–2004 influenza season were identified from virology laboratory reports, and their charts were reviewed.
RESULTS. There were 505 children admitted because of influenza. Fifty-seven percent were <2 years old. Previously healthy children accounted for 58% of all of the cases. Pulmonary and neurologic disorders were the most common underlying chronic conditions. Fever and cough were the most frequent manifestations. Seizures occurred in 9% of cases. Serious complications included myocarditis (2), encephalopathy (6), and meningitis (1). There were 3 influenza-related deaths. Mean duration of hospitalization was 5.3 days. Twelve percent of children required ICU admission, and 6% required mechanical ventilation. Antibiotic therapy was administered in 77% of cases, and 7% received anti-influenza drugs. Information on influenza vaccination was available for 84 of 154 children identified as vaccine candidates. Twenty two had received vaccine, but only 7 children had been fully immunized >14 days before the onset of illness.
CONCLUSIONS. Healthy young children and children with chronic conditions are at risk for serious illness with influenza. Ongoing surveillance is needed to evaluate the impact of changing immunization recommendations on the burden of influenza illness in children.
Key Words: influenza • vaccine • pediatrics
Abbreviations: IMPACT—Immunization Monitoring Program Active • OR—odds ratio • CI—confidence interval
Influenza is a common childhood disease, with attack rates of 20% to 30% per year in preschool children.1–3 It has been recognized recently that risk of hospitalization of healthy young children during influenza season is similar to that of older high-risk individuals and the elderly.4–6 There is less information on the impact of confirmed influenza infections in children. In a Tennessee cohort followed between 1974 and 1999, average annual admission rate for culture-proven influenza was 3.5 per 1000 children <2 years of age.3 More recently, admission rates of 4.3 per 1000 children <6 months old and 0.9 per 1000 children 6 to 23 months old were reported by the New Vaccine Surveillance Network of the Centers for Disease Control and Prevention.7
Children with influenza are typically admitted with respiratory distress, pneumonia, croup, or bronchiolitis. In a study from Texas, influenza was the most common virus identified among children aged 5 to 17 years with chronic conditions hospitalized for respiratory illness; for children <5 years old it was second only to respiratory syncytial virus.8Young infants may present with fever without other findings.9,10 Influenza is frequently associated with febrile seizures.9–11 Encephalitis and acute encephalopathy are uncommon but serious complications.12–14 Influenza and its complications lead to a 10% to 30% increase in the number of antimicrobial courses prescribed to children during influenza season.5
In 2003, the American Advisory Committee on Immunization Practices encouraged influenza vaccination for all children 6 to 23 months old and close contacts of those <2 years of age.15 In anticipation of similar recommendations in Canada, surveillance for hospitalizations for influenza was conducted during the winter of 2003–2004 to obtain baseline data that would permit the evaluation of changing immunization policies. This report describes the characteristics of a large sample of children hospitalized for laboratory-confirmed influenza before expanded immunization policies were implemented. The active nature of the surveillance promoted complete case ascertainment and detailed reporting of risk factors, disease manifestations and sequelae, and resource use. It reaffirms that children who are hospitalized are frequently otherwise healthy and that children with chronic neurologic disease are a high-risk group.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
At each hospital, viral diagnostic tests were routinely performed on children hospitalized with acute respiratory infection. Virology laboratory reports were reviewed to identify children from whom influenza virus had been detected by viral culture or by direct immunofluorescence assay for viral antigen. Hospital records were reviewed by trained nurse monitors or infectious diseases physicians to determine whether those with positive influenza tests had been admitted to the hospital because of influenza. Admission records were also reviewed to identify children with influenza diagnosed in another laboratory. Demographic and clinical data were extracted from the records. Fever was defined as temperature of 
38.5°C (rectal), 38°C (oral, tympanic), or 37.5°C (axillary); respiratory distress as increased respiratory effort, dyspnoea, or oxygen saturation <95%; encephalopathy as change in state of consciousness; or meningitis as an abnormal number of white blood cells in cerebrospinal fluid. Wheezing included bronchospasm, bronchiolitis, and asthma. Pneumonia was recorded only if confirmed by chest radiograph. Other clinical manifestations were recorded as documented in the patient record. Secondary bacterial infections were defined as those with positive bacterial cultures from relevant clinical specimens or a physician's diagnosis of bacterial infection.
Patient anonymity was preserved. Case reports were prepared using a standard data collection form and forwarded to the IMPACT data center in Vancouver, British Columbia, Canada, where data were entered into a database (Access, Microsoft Canada, Mississauga, Ontario, Canada) and transferred to Epi-Info 6.04d (Centers for Disease Control and Prevention, Atlanta, GA) and SAS 8.0 (SAS Institute, Cary, NC) for analysis. Means were compared using Kruskal-Wallis analysis of variance and proportions by Mantel-Haenszel 
2. Influence of age and underlying illness on admission to ICUs were assessed using logistic regression (SAS 8.2, SAS Institute). The project was approved by research ethics boards of participating hospitals.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The number of cases reported by individual participating centers varied from 20 to 92. Cases occurred between September 22 and June 8. The epidemic started in Edmonton, Alberta, Canada, then moved west and east across the country with peak activity in November in Edmonton and Winnipeg, Manitoba, Canada, and from December to February elsewhere (Fig 1).
|
|
A total of 293 children (58%) were previously healthy. The proportion who were healthy decreased with age (2 for linear trend: P < .00001; Fig 3). The 212 children who were not previously healthy had 291 chronic conditions recorded. Pulmonary disease was most frequent, occurring in 18% of all children, followed by neurologic disease in 12% (Table 1). Pulmonary diseases included moderate or severe asthma (35), bronchopulmonary dysplasia (13), cystic fibrosis (4), other chronic lung disease (17), chronic or recurrent aspiration (11), tracheal or laryngeal abnormalities (7), and other pulmonary conditions (3). Neurologic diseases reported were developmental delay with seizures, hydrocephalus, microcephaly, or neuromuscular abnormality (18); seizure disorders (12); cerebral palsy (10); chronic encephalopathy (6); neuromuscular disorders (6); hydrocephalus (3); developmental delay without other abnormality (3); and pseudotumor cerebri (1). Two of the children with neurologic disease were <6 months old, 17 were 6 to 23 months old, 40 were 2 years of age, and 26 (44%) had other significant comorbidities.
|
|
5 years, respectively (2 for linear trend: P < .00001). Gestational age was recorded for 227 of 288 children <24 months of age and was <27 weeks for 8 cases, 27 to 32 weeks for 14, 33 to 36 weeks for 25, and >36 weeks for 180. Of 47 children with gestational age <37 weeks, 64% had an underlying illness, most commonly bronchopulmonary dysplasia.
Clinical Manifestations
Clinical manifestations by age group are shown in Table 2. Fever and cough were the most frequent findings. Cough and pneumonia were less frequent in those <6 months of age, whereas rhinorrhea and dehydration were more frequent in this age group. Wheezing and otitis media were most frequent in those aged 6 to 24 months. Pharyngitis was most frequent in older children. Forty-five children, 30 of whom were previously healthy, had seizures.
|
|
Serious complications included 6 cases of encephalopathy (1.2%), all but 1 in children 6 to 23 months of age; 1 case of aseptic meningitis in a 3-month-old; 2 cases of myocarditis in children aged 3 and 7 years, and 1 case of necrotizing fasciitis in a 7-year-old. Except for 1 case of encephalopathy, all occurred in previously healthy children. There were 3 influenza-related deaths: pneumonia and cardiac arrest in an adolescent with severe congenital heart disease who had received influenza vaccine 7 days before onset of illness, acute myocarditis in a previously healthy 7-year-old, and pneumonia followed by septic shock in a 1-year-old with a mild cardiac malformation. The latter 2 children had not been vaccinated.
Resource Use
Patients were hospitalized for a mean of 5.3 days, median of 3 days, and a total of 2645 days. Requirements for supplemental oxygen, admission to ICU, and mechanical ventilation are shown in Table 4. Two children, 1 previously healthy, required extracorporeal membrane oxygenation for 4 and 14 days respectively.
|
|
The percentage of children requiring admission to an ICU was higher for those 5 years of age than for younger children (21.9% vs 9.9%; OR: 2.55; 95% CI: 1.32–4.90) but did not differ significantly between previously healthy children and those with underlying illness (10% vs 15%). Mean duration of ICU admission was longer for children with underlying illness than for previously healthy children (Fig 4). For children without underlying illness, the duration of ICU admission was longer for children <6 months of age than for older children (4.75 vs 2.76 days; P = 0.04). There were no other significant differences in the duration of ICU admission by age group. The percentage of all of the hospitalized children requiring mechanical ventilation was higher for those 5 years of age than for younger children (12.2% vs 5.0%; OR: 2.66; 95% CI: 1.11–6.24) but did not differ significantly between previously healthy children and those with underlying illness (5.5% vs 7.1%). The mean duration of mechanical ventilation did not differ significantly by health status or age group. In logistic regression using underlying illness and age group as covariates, age >5 years remained an independent risk factor for ICU admission after adjustment for underlying illness (OR: 2.347; 95% CI: 1.21–4.57).
Systemic antibiotic therapy was administered in 390 cases (77%). Secondary bacterial infection was identified or suspected in only 115 (29%) of these. There were no significant differences in the percentages treated by age group or by underlying health status. The percentages treated by center varied between 68% and 95% and did not correlate with the numbers of bacterial infections suspected or diagnosed.
Of those treated, 340 (87%) received parenteral antibiotics with or without oral therapy, whereas 49 (13%) received only oral antibiotics. The route of administration was not documented for 1 case. Mean duration of treatment in hospital was 3.9 days. Eighty-four children (17%) were discharged from the hospital with antibiotics. A single antibiotic was administered in 37% of cases, 33% received 2, and 30% received 3. The antibiotics most frequently used and the numbers of courses were: cefuroxime (181), amoxicillin (80), cefotaxime (73), ampicillin (66), ceftriaxone (50), clarithromycin (41), cefprozil (40), and gentamicin (39). Ten children received acyclovir for herpes simplex orolabial reactivation or suspected herpes encephalitis.
Only 33 patients (7%) received anti-influenza drugs (amandatine [14], oseltamivir [18], or both [1]). Twenty nine of these were hospitalized at 3 centers. Twenty six had underlying disease, and 21 of these were immunocompromised. The 7 previously healthy children all had serious disease: encephalopathy (4), pneumonia requiring ICU care (2), and myocarditis with shock (1).
Influenza Vaccination Status
There were 154 children aged 6 months who were identified as influenza vaccine candidates. Of these, 138 met the Canada National Advisory Committee on Immunization 2003–2004 criteria for specific high-risk groups.17 Another 16 had other chronic conditions (Table 5).
|
Of the 57 children with neurologic disease who were 6 months of age, 26 were considered as vaccine candidates because of comorbidities,14 were considered as vaccine candidates based on neurologic disease alone, and 17 were not considered as candidates. Twenty children aged 2 years had no comorbidities for which influenza vaccine was recommended. Inclusion of neurologic disease alone as a criterion for vaccination varied by IMPACT center. Only 2 children with neurologic disease were fully vaccinated before influenza. Four were incompletely vaccinated, 27 were not vaccinated, and no information was available for 26. The most frequent reasons documented for failure to vaccinate high-risk children were infection occurring too early in the season and parents' lack of awareness of the recommendation for vaccination.
Influenza B
The 5 cases of influenza B occurred in December (1), March (1), and April (3). Ages were 1 to 12 years. Four children had underlying illnesses, including 2 with seizure disorders who were admitted with increased seizure activity. The duration of hospitalization was 3 to 34 days. Two required ICU admission and ventilation.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
16250 children per 3-month period (D. Scheifele, MD, written communication, August 5, 2004). The minimal proportion of admissions attributable to influenza during influenza season is, thus, estimated to be 3%, which is consistent with the 2.2% reported in a pilot study at 1 center.10 The data presented likely underestimate the impact of hospitalization for influenza, because children admitted because of late complications of influenza may not have been tested for the virus or may have had negative test results. Also, although all of the centers reported routine virology testing of children admitted with respiratory infections in influenza season, some young infants presenting with fever without respiratory symptoms may not have been tested. As predicted from other reports,4,5,10 healthy children <2 years of age and children of all ages with underlying chronic illnesses made up the majority of admissions for influenza. Healthy young infants with fever without an obvious focus may be admitted for investigation and empiric therapy of suspected bacterial sepsis.10 Of the chronic underlying illnesses reported, neurologic disorder was second in frequency only to pulmonary disease and resulted in admission because of respiratory failure, increased frequency of seizures, or both. Although neurologic illness is not usually mentioned in reports of chronic conditions predisposing to severe influenza, it was an important underlying disease in a series from Finland.9 A recent study showed neurologic disease to be an important risk factor for respiratory failure in children with influenza.18 Chronic neurologic disorders were the most frequent underlying conditions reported in a series of influenza-related deaths in the United States in 2003–2004.19
The observation that older children were more at risk for ICU admission, even after adjustment for underlying illness, was unexpected. This may be an incidental finding, because the number of older children admitted was low. One possible explanation is that older healthy children were unlikely to be admitted unless they had severe influenza illness, whereas younger children may have been admitted with less severe illness.
A disproportionate representation of Aboriginal children (First Nations, Inuit, or other) has already been reported from one hospital participating in our surveillance,20 where Aboriginal children accounted for 45% of influenza admissions but 8.7% of the pediatric population of the province (Statistics Canada 2001; www12.statcan.ca/english/census01/home/index.cfm). This apparent increase in risk may reflect local referral patterns or socioeconomic or genetic factors and warrants further exploration. Similar overrepresentation was not observed at other IMPACT centers that reported ethnic origin.
Clinical manifestations were mainly respiratory. Seizures occurred in 9%. Although influenza is often a benign disease, serious illness may occur.9–11,14,21 Influenza-related mortality rate in our series of hospitalized children was 0.6%, with 1 death occurring in a previously healthy child. Another 1.8% had serious complications including encephalopathy, meningitis, and myocarditis. Encephalopathy and encephalitis presenting as an altered state of consciousness with or without convulsions are being increasingly recognized in young children with influenza.12,13,22
Although influenza is known to predispose to Staphylococcus aureus, pneumococcal, and meningococcal infections,14,23–25 in our series, laboratory-documented secondary infection was rare. The ability to detect secondary bacterial infections was limited by the nature of this surveillance program. Invasive procedures to determine the etiology of pneumonia, otitis media, or sinusitis are not often performed in children, and precise clinical definitions of secondary bacterial infections were not provided.
There was very little influenza B activity in Canada in the 2003–2004 season.26 The literature suggests that influenza B occurs in older children and may be less severe than influenza A.9,21 The low number of influenza B cases in our series precluded detection of significant differences between the types by age group or clinical presentation.
Pediatric influenza resulted in considerable use of health care resources. As expected, those with underlying chronic illness required more supportive therapy. Empiric antibiotic therapy was very extensively used. The implication of influenza in excess antibiotic use has been reported elsewhere.5,27 In contrast, antiviral agents active against influenza were rarely used and only in high-risk situations. The reasons for this are not known, but possibilities include rapid improvement with supportive care, delayed diagnosis of influenza, and lack of experience with these agents in young children.
Before 2003, influenza vaccination was recommended only for children with certain chronic conditions predisposing them to serious consequences of influenza. The literature suggests that vaccination rates in high-risk children were low.28 Only 31% of high-risk children admitted for respiratory illness during influenza season in 1999–2000 in Tennessee had been vaccinated.29 Recommendation from a physician has been reported as the most significant factor associated with receiving vaccine.29,30 In our study, information on influenza vaccine was incomplete, with data available for only 55% of high-risk children. This was because of the method of data collection, which was limited to chart review in most centers. Only 26% of high-risk children for whom data were available had received vaccine, and most of these were not fully immunized before the onset of influenza. Influenza virus began circulating early in the fall of 2003,26 and several of these cases occurred in October, when immunization campaigns were just getting underway. Nevertheless, strategies are needed to improve immunization of high-risk children.
Children with underlying neurologic disease have not been specifically mentioned in recommendations for influenza vaccination in the past and, depending on how chronic disease is interpreted by vaccinators, may not be offered vaccine.10 Recommendations for influenza vaccination have been modified recently to include persons with cognitive dysfunction, seizure disorders, or neuromuscular disorders that may compromise respiratory function or handling of respiratory secretions or increase the risk of aspiration as a high-risk group.31,32
In the United States, influenza vaccine was encouraged for all children 6 to 23 months old and household contacts and out-of-home caretakers of children <2 years old for the 2003–2004 influenza season15 and recommended for the 2004–2005 season.33,34 In June 2004, the Canadian National Advisory Committee on Immunization made similar recommendations for 2004–2005.26 It is anticipated that these recommendations will decrease the burden of influenza in children. The impact will be influenced by problems of implementation, especially introduction of another vaccine for infants into an already crowded vaccine schedule and the degree of vaccine acceptance by parents. Despite a publicly funded universal influenza immunization program launched in Ontario, Canada, in the fall of 2000, only 27% of children presenting to a pediatric emergency department in early 2001 had been vaccinated.30 Studies have shown inactivated influenza vaccine to be immunogenic and effective in children, but data specific to children <2 years of age are scarce.31,35–37 Although the effectiveness of vaccinating children <2 years of age has been challenged,38 individual studies have shown effectiveness in this age group.39,40
Children <6 months of age are also prominently represented in our series. Because influenza vaccine has limited immunogenicity in this age group,41 alternative strategies are required, such as passive protection by maternal immunization during pregnancy and immunization of household contacts and caretakers.15,26,34,35
Our study was limited by inadequate documentation of influenza vaccination history and by the lack of population data required to enable the calculation of age-specific infection rates. These issues are being addressed in ongoing surveillance for influenza hospitalization by the IMPACT program.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Recent changes in influenza vaccination policy have made vaccine available to a much larger segment of the pediatric population. Active prospective hospital-based surveillance with attention to details of influenza vaccination history will be important in evaluating the impact of changes in policy and the effectiveness of influenza vaccine in preventing severe influenza in young children.
|
ACKNOWLEDGMENTS |
|---|
We gratefully acknowledge the expert work of the IMPACT nurse monitors, nurse liaison, and the data center staff. We thank microbiologists Dr Yvonne Yau (Department of Laboratory Medicine, Hospital for Sick Children, Toronto), Dr Bonita Lee (Provincial Laboratory for Public Health, Edmonton, Alberta), and Dr Eva Thomas (Microbiology Program, British Columbia Children's Hospital, Vancouver, British Columbia) for valuable assistance with this project. We also thank Dr Caroline Quach (Infectious Diseases Division, Montreal Children's Hospital) for assistance with the data analyses.
|
FOOTNOTES |
|---|
Address correspondence to Dorothy L. Moore, PhD, MD, Room C1242, Montreal Children's Hospital, 2300 Rue Tupper, Montreal, Quebec, Canada H3H-1P3. E-Mail: dorothy.moore{at}muhc.mcgill.ca
Financial Disclosure: Dr Halperin has received grants and contracts for clinical trials with manufacturers of influenza vaccines (ID Biomedical, GlaxoSmithKline, and Sanofi-Pasteur) and has served on advisory boards for those companies on issues unrelated to influenza vaccine. Dr Vaudry has received an honorarium from Sanofi Pasteur for a speaking engagement. All other authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2006 American Academy of Pediatrics. All rights reserved. |
||
|
||
