ARCHIVED - Active surveillance for influenza infection in children; Stollery Children's Hospital, 2003-2004 season

 

Introduction

Recommendations have been made nationally(1) and internationally(2) to expand routine influenza immunization to all children 6 - 23 months of age. In anticipation of this recommendation, we initiated active surveillance for influenza infection at the Stollery Children's Hospital (SCH) in Edmonton, Alberta for the 2003-2004 influenza season funded by Alberta Health and Wellness (AHW).

Methods

Active prospective surveillance for influenza infection in children < 17 years of age was performed at the SCH from 14 October, 2003 and is ongoing. Surveillance was performed through the exisitng Canadian Pediatric Society/Health Canada Immunization Monitoring Program, Active (IMPACT) unit at the SCH. The study was reviewed and approved by the Health Research Ethics Board of the University of Alberta. Cases were identified by a combination of methods.

  1. Laboratory-based surveillance: All specimens positive for influenza by Direct Fluorescent Antibody test or rapid culture in the Provincial Reference Virology Laboratory originating from the SCH Emergency, clinic or inpatient wards were identified. The date of the case was defined as the day the diagnostic test was positive.

  2. Infection control: The cases identified by infection control were monitored. Any additional cases were added to the surveillance.

  3. Review of daily hospital admission: Any admitting diagnosis clinically consistent with influenza was reviewed. Cases with influenza confirmed in another laboratory were added to the surveillance.

  4. Daily verification with ward nurses and medical staff including the infectious diseases consultant.

To be included in the surveillance, the influenza infection had to be the reason for admission or significantly complicate the underlying condition resulting in admission or be acquired nosocomially. Cases where influenza was diagnosed incidentally were not included.

A case report form was designed to collect data on: demographics, underlying medical conditions, influenza immunization history, clinical manifestations, reason for admission, antiviral therapy, complications, antibiotic therapy, level of care required and travel. Data were collected by a combination of chart review and parental interview. Determination of secondary bacterial infection was based on clinical manifestations of the infection. In the case of pneumonia, this was categorized by the investigator as a secondary bacterial infection if there were clinical, radiographic or laboratory characteristics of secondary bacterial infection. All cases of pneumonia were not considered bacterial; some were considered manifestations of the primary influenza infection.

Most of the in-patient cases were identified prospectively as they were admitted to hospital which allowed for more complete data collection by parental interview regarding ethnicity, travel history and influenza immunization. This information was not reliably documented in the medical record. Immunization history was confirmed either by public health record or by parental history in those who were eligible for influenza vaccine according to then current AHW and National Advisory Committee on Immunization (NACI) guidelines(3). Reasons for not administering vaccine were explored when possible. Cases in the first 2 weeks of October were identified retrospectively so travel and immunization data are not as complete. For nosocomial cases the charts were reviewed and data collected for the duration of hospitalization after and during the manifestations of influenza infection or its complications.

Regular reports summarizing the total number of cases and identifying significant morbidity were provided to AHW during the surveillance. Differences between groups were assessed using the Student's t-test, chi-squared test and logistic regression.

Results

One hundred and fifty-nine cases of influenza infection were identified at the SCH; 93 of these were in-patients of which 90 were admitted because of influenza and 3 developed infection nosocomially; 66 were out-patients, mostly in emergency.

Epidemiology

The outbreak began very early in Edmonton with the first cases identified at the SCH on 2 October, 2003. The last case was identified on 4 January, 2004. The outbreak continued at a high rate from mid October to mid November with more than 20 cases identified per week for that 4-week period (Figure 1).



Figure 1. Influenza Outbreak, Stollery Children's Hospital , October 2003 - January 2004, Weekly Case Totals

Figure 1. Influenza Outbreak, Stollery Children's Hospital , October 2003 - January 2004, Weekly Case Totals


Virology

All isolates were Influenza A and those that were characterized were the H3N2 variant, Fujian strain.

Demographics

  1. Age and health status: The mean age of all children identified was 3.1 years and 55% were < 2 years of age. Sixty-one percent of children admitted to hospital were < 2 years of age and 23% were < 6 months. For children admitted to hospital, 60% of children < 2 were otherwise healthy. The children < 2 were further analyzed according to age: < 6 months (currently not vaccine candidates) and age 6-23 months (Table 1). Younger children admitted to hospital were more likely to be otherwise healthy: 90% of infants < 6 months; 42% of children 6-23 months; and 19% of children> 2 were otherwise healthy (p < 0.001). The most common underlying conditions identified were pulmonary and neurologic. The gestational age was determined for 77 of the cases; 21 (27%) were < 37 weeks gestation. Prematurity was most important as a risk factor for children 6-23 months of age. Immune deficiency was rare as a risk factor in children < 2. In 72 older children, it was present in 8 (11%) of those diagnosed, being equally distributed between those admitted and not admitted.

  2. Ethnicity: Determining ethnicity was not always possible and was more difficult for out-patients where the emergency charts were usually reviewed after the patients were discharged. Of the 93 children admitted, the ethnicity was determined for 65. Thirty children were Aboriginal (10 First Nations, 1 Inuit, 19 other Aboriginal groups). This represented 46% of children whose ethnicity was known and 32% of all admitted children. Twenty-nine (45% of children with known ethnicity, 31% of all admitted children) were Caucasian. Of admitted children < 2 years old, 25 were Aboriginal (64% of children with known ethnicity, 42% of all admitted children < 2).

Table 1. Influenza Cases by Age and Health Status, Stollery Children's Hospital, 2003-2004 Season

HOSP

< 6 months, n(%)

6-23 months, n(%)

> 2 years, n(%)

All ages, n(%)

STATUS

H

NH

Total

H

NH

Total

H

NH

Total

H

NH

Total

Admitted

19(90)

2(10)

21

15(42)

21(38)

36

7(19)

29(81)

36

41(44)

52(56)

93

Not Admitted

4(80)

1(20)

5

19(76)

6(24)

25

20(56)

16(44)

36

43(65)

23(35)

66

Total

23(88)

3(12)

26

34(56)

27(44)

61

27(38)

45(62)

72

84(53)

75(47)

159

H = Healthy; NH = Not Healthy

 

Clinical Manifestations

The most common symptom was fever, which was present in 94% of children overall. Ten percent of children had lethargy, 55% of children presented with rhinorrhea and 88% of patients had cough. Pharyngitis and otitis media were each diagnosed in 10% of children. Respiratory distress was present in 23% of cases; 15% were diagnosed with pneumonia, 5% had croup and none had apnea. Nausea and vomiting (34%), diarrhea (16%) and dehydration (18%) were common. Seizures were present in 8% of children; either simple febrile seizures or recurrent seizures in children with a pre-existing seizure disorder. There was one case of encephalitis and two cases of meningitis.

Secondary Bacterial Infection

Secondary bacterial infection was diagnosed in 16 (10%) children overall, and 13% of in-patient cases. There were 5 cases of pneumonia, 2 bacteremias (Streptococcus pneumoniae and Pseudomonas aeruginosa), 2 urinary tract infections, 2 episodes of cellulitis and 5 other diagnoses including staphylococcal scalded skin syndrome and mastoiditis.

Antimicrobial Utilization

More than half (58%) of all children identified received antibiotic therapy. There was a trend for antibiotic usage to be more common in younger children (< 6 months, 65%; 6-23 months, 49%; > 2 years, 40%) but this was not statistically significant. However, antibiotic usage was significantly more common among admitted (82%) than not admitted (25%) children (p < 0.001). Four children were treated with oseltamivir. No other antiviral therapy was used.

Worsening of Underlying Disease

Overall, the underlying disease had worsened in 38 (24%) of the children. The most common disease state that deteriorated was asthma with 15 children (39%) reporting increased wheezing with another 7 (18%) reporting increased supplemental oxygen or ventilator support required. Six children (16%) with underlying seizure disorders reported increased frequency of seizures.

Use of Hospital Resources

  1. Admitted patients: A total of 435 hospital days were attributable to influenza: < 6 months: 72 days; 6-23 months, 168 days; > 2 years, 195 days. Forty-seven children (51%) required supplemental oxygen. Five children required mechanical ventilator support. There was a trend for younger children to have a shorter admission time and fewer days of oxygen therapy but this was not statistically significant.

  2. Nosocomial cases: There were three nosocomial cases. One of these three cases was a 7-month-old child who was in hospital for chemotherapy and tested positive for influenza on the day of discharge. The child was readmitted within 24 hours with worsening respiratory distress and was admitted to the Pediatric Intensive Care Unit (PICU) with S. pneumoniae septic shock. The infant required intensive care for 11 days, required ventilatory support for 8 days and supplemental oxygen for 22 of the 29 days in hospital.

  3. ICU admission: Seven children were admitted to the PICU for a total of 60 days (range 1-11 days). One infant was < 6 months, 3 were between 6 and 23 months, and 3 were > 2 years of age.

Outcome

There were no deaths identified due to influenza. All in-patients were recovering and discharged or still hospitalized for another reason. One child was improving and transferred to another facility (although not fully recovered from encephalitis).

Risk Factor/Indication for Vaccine

Sixty children had risk factors that made them candidates for vaccine administration according to 2003 AHW and NACI guidelines(3). This represented 38% of the children; 42% of admitted patients and 24% of not admitted patients. The most common risk factor was pulmonary disease, 47%. The next most common risk category (22%) was other chronic medical condition not specified in the NACI guidelines. The most common "other chronic medical condition" identified was neurologic. Ten children did receive the vaccine and 7 of these were admitted to hospital. All of these children developed symptoms within 14 days of vaccine administration; there were no documented vaccine failures.

Reasons for not immunizing

Of the 60 children eligible for vaccine: 32% of the cases occurred too early in the season for the vaccine to have been effective (i.e. within 14 days of vaccine becoming available in the community on 14 October); 20% of the families were unaware that their child was a vaccine candidate; 2 children were allergic to eggs, 1 was too young; and 1 child's family had made an appointment to receive vaccine but the child developed infection before the appointment could be kept (this child was admitted to the PICU).

Travel

Travel history could not be obtained in 60% of reported cases. Where the travel history was obtained, family members of 12 children (19%) had travelled within the last 30 days: 3 within the home province, 4 to other parts of Canada, 1 to the continental U.S.A., 1 to Asia (China), 1 to Australia, 1 to Central America (El Salvador), and 1 not determined.

Discussion

This paper reports the first active prospective hospital-based surveillance for influenza infection in children. The active and prospective nature of the data collection allow for complete description of the demographics, risk factors, immunization history, clinical manifestations, complications and required hospital resources associated with influenza infection in children at a tertiary-care referral children's hospital in Canada. Because this surveillance occurred with the first outbreak of the Fujian strain of influenza, a complete, timely description of the clinical manifestations in severely ill children was possible.

As has been previously described(4-6), there is a disproportionate burden of illness from influenza in young and otherwise healthy children. The majority of children hospitalized were < 2 years and a large proportion was otherwise healthy. A significant burden of illness and hospitalization was also seen in children < 6 months of age who were currently not candidates for vaccine. The population of hospitalized children was slightly older in this study than in a retrospective study from the Montreal Children's Hospital(6) (3.3 years vs 26 months) and the proportion of infants < 6 months of age was also lower (23% vs 34%). This may reflect a difference in the clinical manifestations of the new Fujian strain or differences in referral patterns for the respective hospitals. Aboriginal children were also disproportionately represented in this study. Aboriginal people represent approximately 3% of the population of Alberta and yet we observed the same number of hospitalizations for Aboriginal children as for Caucasian children. This may represent a genetic predisposition to more severe infection or be related to socioeconomic conditions resulting in a higher risk of infection, higher likelihood of complications or a greater likelihood of being hospitalized.

The clinical manifestations of influenza infection were as expected. The most common presentations were respiratory distress, lethargy, dehydration and seizures. The important association between seizures and influenza infection has been noted by others(7). There was one case of encephalitis associated with acute infection in the respiratory tract; however, no viral diagnostic testing of cerebral spinal fluid or brain tissue was performed. Although the child did recover, there were residual neurologic sequelae at the time of transfer. Severe encephalopathy has been described after influenza infection, particularly in Japan(8), but also in Canada(9) and the United States(10).

Bacterial coinfection occurred in an important number of children. The diagnosis of bacterial pneumonia is difficult in young children and may not have been accurate as it was based on radiographic and clinical parameters rather than microbiologic proof. However, an effort was made to separate pneumonia occurring as a consequence of influenza from that which clinically appeared to be a bacterial suprainfection as described in the methods. Two episodes of bacteremia were related to influenza infection. One immunocompromised child developed influenza nosocomially then presented within 24 hours with S. pneumonia septic shock and was admitted to the PICU for a total of 11 days. The other child with combined immune deficiency and an indwelling catheter developed P. aeruginosa bacteremia after being diagnosed with influenza and responded to antibiotic therapy. Two episodes of urinary tract infection were diagnosed in children with structural urinary tract anomalies and recurrent urinary tract infections.

Empiric antibiotic use was much more common than the diagnosis of bacterial infection, which likely reflects the difficulty for the clinician in defining viral versus bacterial causes of symptomatology, as well as a recognition of the risk of secondary bacterial infection with influenza. There was very little antiviral use and it was exclusively oseltamivir.

Many children had worsening of their underlying disease. The most common underlying condition that deteriorated was respiratory disease, particularly asthma. This increased burden of illness from influenza infection in children with asthma has been documented previously(11). In this study some children with asthma, even clinically mild asthma on no chronic regular medications, required hospitalization for significant wheezing and shortness of breath. Children with underlying neurologic disease comprised the next most common category presenting with deterioration of their underlying condition, frequently with increased seizures.

A total of 93 children with an acute virologic diagnosis of influenza infection spent over 400 days in a tertiary-care children's hospital with 60 of those days spent in the PICU. For many of those days, children required supplemental oxygen, mechanical ventilation and intravenous antibiotics. It is important to realize that these cases of influenza infection, diagnosed virologically at the time of hospital admission, probably only reflect a fraction of the disease burden from influenza in children. Much of the morbidity and possible mortality occurs as a later complication of influenza and there is no laboratory-proven diagnosis of antecedent influenza infection. Population studies of disease rates during influenza season are more reflective of the magnitude of the problem(4-5). A recognition of the cost of providing care for children with influenza may help to justify the cost of expanding vaccination programs as recommended in the latest NACI statement(9). Nosocomial influenza resulted in significant morbidity in this study and reiterates the importance of at-risk children being immunized, including those in hospital.

Because of the timing of the outbreak in Edmonton in 2003, it is difficult to judge adequately the effectiveness of the vaccine program because many of the children in the study were infected before they could have had accessed vaccine that year. There were no documented vaccine failures although this could have been because the outbreak was almost over before much of the population was immunized. Interestingly, 20% of the parents of the at-risk children hospitalized with influenza, did not realize their child was eligible for vaccine. This suggests an ongoing need to educate caregivers of the need for influenza vaccine in this high-risk population.

Obtaining travel history was feasible but was somewhat difficult to interpret in this study. However, active real-time prospective surveillance could be used to collect timely travel histories in a remote setting where there is an outbreak of respiratory disease.

Conclusion and Recommedations

  1. Influenza immunization should be provided for all children 6-23 months of age and their household contacts.

  2. Household contacts of infants < 6 months should be immunized in order to provide a protective "cocoon" to these high-risk infants.

  3. Further promotion of influenza immunization should occur for high-risk children and their families.

  4. Certain risk groups need to be addressed in immunization programs:

    1. Children with underlying neurologic disease

    2. Children with asthma, even those with mild disease

    3. Aboriginal children

  5. Active prospective surveillance of hospitalizations in children can provide reliable, timely monitoring of severe disease and should be part of a comprehensive influenza surveillance program.

Acknowledgements

This study was supported by a grant from AHW. The authors thank Dr. D. Moore, Montreal Children's Hospital, Montreal, Quebec, for comments on the case report form; B. Leblanc, Medical Laboratory Technologist, Provincial Laboratory for Public Health, Edmonton, Alberta; and staff at the Vaccine Evaluation Center, Vancouver, British Columbia, for data management.

References

  1. National Advisory Committee on Immunization. Statement on influenza vaccination for the 2004-2005 season. CCDR 2004;30:1-32.

  2. American Academy of Pediatrics Policy Statement. Recommendations for influenza immunization of children. Pediatrics 2004;113(5):1441-46.

  3. National Advisory Committee on Immunization. Statement on influenza vaccination for the 2003-2004 season. CCDR 2003;29:1-20.

  4. Neuzil KM, Mellen BG, Wright PF et al. The effect of influenza on hospitalizations, outpatient visits, and courses of antibiotics in children. N Engl J Med 2000;342(4):225-31.

  5. Izurieta HS, Thompson WW, Kramarz P et al. Influenza and the rates of hospitalization for respiratory disease among infants and young children. N Engl J Med 2000;342(4):232-39.

  6. Quach C, Piche-Walker L, Platt R et al. Risk factors associated with severe influenza infections in childhood: implication for vaccine strategy. Pediatrics 2003;112(3):e197-201.

  7. Chiu SS, Tse CY, Lau YL et al. Influenza A infection is an important cause of febrile seizures. Pediatrics 2001;108(4):E63.

  8. Sugaya N. Influenza-associated encephalopathy in Japan. Semin Pediatr Infect Dis 2002;13:79-84.

  9. Sazgar M, Robinson JL, Chan AKJ et al. Influenza B acute necrotizing encephalopathy: a case report and literature review. Pediatr Neurol 2003;28:396-99.

  10. Weitkamp JH, Spring MD, Brogan T et al. Influenza A virus-associated acute necrotizing encephalopathy in the United States. Pediatr Infect Dis J 2004;23:253-54.

  11. Neuzil KM, Wright PF, Mitchel EF et al. The burden of influenza illness in children with asthma and other chronic medical conditions. J Pediatr 2000;137:856-64.

Source: W Vaudry, MD, CM, FRCP (C), Division of Infectious Diseases, Department of Pediatrics, University of Alberta, Stollery Children's Hospital, Edmonton; A Roth, RN, Division of Infectious Diseases, Department of Pediatrics, University of Alberta, Stollery Children's Hospital, Edmonton; B Lee, MD, FRCP (C), Division of Infectious Diseases, Department of Pediatrics, University of Alberta, Stollery Children's Hospital, Provincial Laboratory of Public Health (Microbiology), Edmonton; D Spady, MD, MSc, Department of Pediatrics, University of Alberta, Stollery Children's Hospital, Edmonton, Alberta.

 

Page details

Date modified: