Objective To determine the clinical characteristics of infants ≤2 months old hospitalised with influenza A.
Study design The study was a retrospective analysis of infants ≤2 months old hospitalised with fever, respiratory distress and/or sepsis. Clinical signs, laboratory values, hospital duration and outcome were compared between children with influenza A and other viruses.
Results The charts of 268 infants were reviewed. 29 (11%) children had laboratory-confirmed influenza A infection. Unique features associated with influenza infection included the high number presenting with fever (97%) and a history of exposure to family members with a flu-like illness (69%). A significantly lower rate of respiratory distress was observed in the influenza group compared with respiratory syncytial virus (RSV) (24% vs 89%, p≤0.001). Median duration of hospitalisation for influenza was shorter than RSV (4 days vs 6 days, p<0.001).
Conclusions In young infants, influenza A is a relatively mild disease compared to RSV and is primarily associated with upper respiratory tract manifestations.
Statistics from Altmetric.com
Influenza A infection during infancy is a significant burden for the health system despite the increased rate of vaccination.1 Infants infected during the first year of life have a higher rate of hospital admissions, a greater number of complications compared with children of other age groups and adults, and a higher mortality rate. With regard to the first 2 months of life, very few studies evaluated the severity of influenza infections as these infants may be relatively protected through transfer of maternal antibodies.2 In the present study we aimed to study the clinical characteristics of influenza A infections in infants ≤2 months and compare them with infants with other respiratory viruses, including respiratory syncytial virus (RSV).
The study was conducted at the Paediatric Departments of the Hadassah Medical Center, Jerusalem during the calendar year 2003. To be eligible the child had to be 2 months of age or younger (≤62 days), and be hospitalised for either: (1) acute respiratory disease defined as tachypnoea, hypoxia or respiratory distress; (2) acute life-threatening event; or (3) clinical suspicion of sepsis based on fever, changes in mental status or other manifestations associated with bacterial infections. Demographic and clinical information was obtained from the chart including whether there was a history of exposure to specific viral pathogens. Viral studies were performed for all children and bacterial/radiology studies when clinically indicated. Nasal wash specimens were analysed for the presence of RSV, influenza A and B, parainfluenza and adenovirus by direct immunofluorescence assay. Antigenic typing and subtyping of influenza virus isolates was performed by haemagglutination inhibition. The study was approved by and performed in adherence to the guidelines of the Hadassah-Hebrew University Medical Center Institutional Review Board. The data was analysed using both parametric and non-parametric statistics. For non-parametric variables, a χ2 or Fisher's exact test was used. Interval-ratio data were compared using the Student's t test, and the Mann–Whitney rank sum test was used for comparison of data that were not normally distributed.
During the course of the study, 268 hospitalised children meeting the study eligibility criteria were evaluated. Of these, 29% had a background illness and 27% had a history of exposure to an individual with a viral illness. The microbiologic distribution of pathogens is shown in figure 1. For a majority (55%), no pathogen was identified, for a minority (7.5%), bacteria could be isolated and for the remainder (38%) either a viral or combined viral/bacterial source for infection was determined. For all 101 children with positive viral tests, RSV was the major pathogen detected (n=53) followed by influenza A (n=29). Subtyping of influenza isolates showed that they were identical to A/Fujian/411/02-like viruses circulating during the latter half of 2003.
Clinical, laboratory, radiological and outcome variables of influenza A infection were compared to other viral infections (RSV, adenovirus, parainfluenza), in general, and RSV infection, in particular (table 1).
Over two-thirds of children with influenza A had a history of exposure to a family member with upper respiratory symptoms; in contrast, for RSV only 20% and for other viruses only 44% had a history of exposure (influenza vs RSV, p≤0.001). A slight male dominance was observed in children hospitalised with influenza infections. Fever >38°C was a nearly invariable finding among children with influenza A infections (97%), but was only found in 46% with RSV (influenza vs RSV, p≤0.001). Although a higher percentage of children had central nervous system manifestations (sleepiness or excessive irritability) in the influenza group compared with RSV (24% vs 9%), this was not significant.
A significant number of children with influenza A had bacterial co-infections. Blood cultures and urinalyses were performed for all patients with influenza infection, and for 21 children, cerebrospinal fluid specimens were obtained as well. Three of 29 children (10%) had a concurrent urinary tract infection associated with positive cultures for bacterial pathogens (two with Group B Streptococcus and one with Staphylococcus aureus), in contrast to RSV infections where only 3/54 (6%) had a concurrent serious bacterial infection. Three other children had acute otitis media. A small number of children with influenza infection (14%) had a background illness including gastro-oesophageal reflux, history of urinary tract infection, congenital stridor and failure to thrive.
Pulmonary involvement of children with influenza A was milder than with RSV or other viruses (table 1). Fewer children with influenza A showed respiratory distress compared with those with RSV (24.1% vs 88.9%, p≤0.001). Similar significant differences were seen for hypoxia. Of the patients who had chest radiographs, none of the children with influenza A had pneumonia (one patient had atelectasis) compared with 17 out of 44 RSV patients. Influenza A patients did have hyperinflation suggestive of reactive airway disease, but this was relatively mild compared with RSV patients, as only a small number of infants with influenza suffered from either hypoxia and/or respiratory distress.
Outcome for patients with influenza based on time to discharge was better as patients with influenza were hospitalised for a median of 4 days compared with 6 days for RSV (p<0.001). For other outcome variables for example, time to defervescence, number of intensive care admissions and invasive pulmonary procedures, no differences were seen between influenza and RSV groups (table 1). No fatalities among the children hospitalised for influenza A occurred and only one child with influenza A, a 28-day-old female, was admitted to the paediatric intensive care unit because of repeat cyanotic events and hypoxia, complicated by persistent gastro-oesophageal reflux.
Influenza A was the second most frequent cause for hospitalisation after RSV in young infants (≤2 months) hospitalised for respiratory distress, fever, apparent life-threatening event or clinical suspicion of sepsis. Compared with RSV infants with influenza A had a milder disease with significantly fewer children having respiratory distress, hypoxia, pneumonia and/or upper respiratory manifestations.
There are several possibilities for the milder course of influenza A compared to RSV. Recent evidence suggests that the incidence and severity of influenza in the first 6 months of life is reduced as a result of maternal immunisation during pregnancy.2 Another possibility is that the immune response to influenza infections is different from that of RSV.3 For example, Welliver et al showed that compared with RSV, infants with influenza had higher levels of lymphocyte-derived cytokines including interleukin (IL)-2, IL-4, interferon γ and IL-17 in nasopharyngeal secretions. More CD-8 positive lymphocytes/natural killer cells as well as a smaller viral load were detected in pathologic specimens from lungs of infants who died of influenza.4
With the introduction of rapid antigen testing, it is now possible to identify influenza infections quickly. Based on the fairly mild course of influenza A in infants and studies showing significantly reduced serious bacterial infections in infants ≥28 days, it has been argued that the management of young infants with influenza infections should be changed in terms of performing less extensive laboratory and radiological studies.5 While our study is supportive of the fact that influenza infections are relatively mild in young infants, over 10% of the infants in our study had a coexisting serious bacterial infection. This would suggest caution before changing the management of febrile young infants with confirmed influenza infections.
The first two authors contributed equally to this work.
Competing interests None.
Ethics approval This study was conducted with the approval of the Hadassah Medical Center Institutional Review Board.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.