Burden of influenza during the first year of life

Abstract Background Every year, influenza viruses infect millions of children and cause an enormous burden of disease. Young children are at the highest risk for influenza‐attributable hospitalizations. Nevertheless, most young children are treated as outpatients, and limited data are available on the burden of influenza in these children. Methods We carried out a prospective cohort study and followed 431 infants born in June‐August 2017 for 10 months from September 1, 2017, to June 30, 2018. The parents filled out daily symptom diaries and were instructed to bring their child for clinical examination at our study clinic each time the child had fever or any signs or symptoms of respiratory tract infection. During each visit, we obtained nasopharyngeal swab specimens for determination of the viral etiology of the illness. Results A total of 55 episodes of laboratory‐confirmed influenza were diagnosed among the 408 actively participating children, which corresponds to an annual incidence rate of 135/1000 children (95% Cl, 102‐175). Excluding five children with double viral infection, acute otitis media developed as a complication of influenza in 23 (46%) children. One (2%) child with influenza was hospitalized because of febrile convulsion. The effectiveness of influenza vaccination was 48% (95% CI, −29%‐80%). Conclusions The burden of influenza during the first year of life is heavy in the outpatient setting where most infants with influenza are managed. Effective strategies for the prevention of influenza particularly in infants under 6 months of age are needed to diminish the burden of disease in this age group.

Influenza-associated childhood mortality rates are also highest among the youngest children. 8,9 Furthermore, children have a central role in the transmission of influenza in the community. 10,11 Prevention of influenza in young infants is difficult. Influenza vaccines are licensed for use only in children older than 6 months of age, and the response to influenza vaccination in young infants is generally weaker than in older children. 12 Maternal influenza vaccination during pregnancy reduces the risk of influenza in infants, but the duration of protection afforded by maternal antibodies is limited. 13,14 Because most infants probably contract influenza from their family members, influenza vaccination of the other family members might provide some protection for infants too young to be vaccinated themselves. 15 In many cases, however, the only way of reducing the burden of influenza in infants is by ameliorating the illness by the use of oseltamivir treatment. [16][17][18] Although the high rates of influenza-associated hospitalization among young infants have been described in several studies, 2-6 most young infants are seen and treated as outpatients. Only limited data exist on the clinical features and overall burden of influenza among outpatient children during their first year of life. 19,20 The aim of this study was to assess the incidence, clinical presentation, and complications of influenza virus infections in a prospectively followed cohort of newborn infants during their first influenza season.

| Study design
This prospective cohort study was performed at a primary care study clinic in Turku, Finland, from September 1, 2017, to June 30, 2018. Before the study period, 431 children born in June-August 2017 at Turku University Hospital were enrolled in a follow-up cohort. The parents of newborn children received written information about the study at the maternal ward of the hospital soon after the child was born, and those who wanted their child to participate signed an informed consent form. Of all children born during the enrollment period, approximately half were enrolled in this study. The study protocol was approved by the Ethics Committee of the Hospital District of Southwest Finland (approval number 47/1801/2017).

| Study conduct
The study clinic was open every day during the 10-month study period, including weekends and holidays. The parents were instructed to bring their child for clinical examination at the study clinic as soon as possible after the onset of fever or any signs of respiratory infection. During each visit, a study physician examined the child, filled out a structured medical record, and obtained nasopharyngeal swab specimens for determination of the viral etiology of the illness. The structured medical record contained detailed questions about the presence and duration of the child´s preceding signs and symptoms and detailed findings at the clinical examination. Oseltamivir treatment was given to all children who were diagnosed with a laboratory-confirmed influenza within 48 hours of illness onset. The dosage of oseltamivir was 3 mg/kg twice daily for 5 days.
The children were routinely re-examined on days 5-7 after the onset of illness and additionally whenever the parents deemed it necessary, especially if they suspected the worsening of the disease or the development of a complication such as acute otitis media (AOM). All visits were free of charge to the families, and there was no limit for the number of visits during the study.
Background information regarding the family, pregnancy, and delivery was obtained from the parents of all children. The parents were provided with symptom diaries (one for September-January and another for February-June) that they were asked to complete daily throughout the 10-month follow-up period. Children were considered active participants if they visited the study clinic at least once or if the parents returned at least one of the two symptom diaries, and if the parents did not inform the study personnel that their child had been treated for a respiratory illness somewhere else than at the study clinic. The baseline characteristics of the 408 active participants are shown in Table 1.

| Definitions
The overall duration of illness consisted of all consecutive days on which the child had fever, rhinitis, or cough. The diagnosis of AOM

| Virologic methods
Two nasopharyngeal flocked swab specimens (Ultra minitip, Copan Italia S.p.a, Italy) were collected for viral analyses from each child at the initial visit for each respiratory infection. One of the specimens was analyzed onsite at the study clinic by rapid antigen     Gastrointestinal symptoms were present in 11 (22.0%) children.

| Clinical presentation at the initial visit
None of the differences between influenza A and B infections were statistically significant.

| Duration of symptoms
The median duration of preceding symptoms before the initial visit

| Complications and management
The most frequent complication of influenza was AOM which was

| D ISCUSS I ON
Acute otitis media 6 (23) 6 (25) 12 (24) representative group of young infants followed carefully throughout their first influenza season. Although virtually all infants in the cohort were taken care of at home, 13% of them contracted a laboratory-confirmed influenza. In almost half of children with influenza, the course of illness was complicated by the development of AOM that was mostly treated with antibiotics. Our study also confirms that the greatest part of the total burden of influenza even among the youngest children occurs in the outpatient setting. Although none of the infants had had prior exposure to influenza viruses and only a minor proportion of them were vaccinated, only one child was hospitalized with influenza.
Our observed incidence rate of influenza during the first year of life is comparable with previous results among older children that have demonstrated annual influenza attack rates of 10%-20% in children. 21,22 In one recent follow-up study, the incidence of laboratory-confirmed influenza among young children during their first influenza season was only 5%. 20 One potential explanation for the difference is in the study design; in that previous study, substantial numbers of nasal swab samples were obtained at home by the parents who mailed the specimens to the laboratory, and eventually, a specimen for viral diagnosis was available for approximately half of all respiratory infections in the cohort. However, it is also wellknown that influenza epidemics vary in intensity and severity, and the attack rates could be lower than average during any milder influenza seasons.
In this study, AOM was diagnosed as a complication of influenza in 46% of young children, which is in agreement with previ- In previous studies, fever has been the strongest sign associated with influenza 20,27-29 and that applied also to the infants in our study. However, as observed also previously, rhinitis and cough were frequently present already in the early phase of the illness. 27,29 This corroborates earlier conclusions that especially in young children influenza virus infections are clinically indistinguishable from other respiratory viral infections, and they cannot be reliably identified without the use of specific diagnostic tests. 30,31 The duration of symptoms before the initial visit to the study clinic was shorter in children with influenza A than in those with influenza B. It is therefore possible that the early symptoms of influenza A infection are more prominent in infants than symptoms caused by influenza B viruses. Although this finding suggests a need for further studies, it is also possible that the observed difference was due to chance only, because ample recent data indicate that there are no differences in the clinical presentation of influenza A and B virus infections in children. [32][33][34][35] Furthermore, no other differences between influenza A and B were observed in this study.
Although the point estimate of the effectiveness of influenza vaccination of infants in our study was 48%, it was not statistically significant. This is in line with the nationwide analysis conducted by the Finnish Institute for Health and Welfare, concluding that no effectiveness could be shown for the trivalent inactivated influenza vaccine used during the season of our study. 36,37 Because vaccine effectiveness was low not only in children but also in adults, it could well explain why vaccination of all other family members in our study did not reduce the incidence of influenza in the infants. In the absence of an influenza vaccine that could be administered to infants under 6 months of age, many health authorities emphasize the importance of vaccinating all household members and other close contacts to protect infants from influenza. 15 It is important to notice that our results do not demonstrate that such a "cocooning" strategy In conclusion, our study provides new and detailed information about influenza in young infants during their first influenza season.
Besides the relatively high rates of influenza-associated hospitalization among the youngest infants, the burden of illness is heavy also in the outpatient setting where most infants with influenza are managed. Because influenza vaccines are licensed only for children older than 6 months and the duration of protection afforded by maternal antibodies is limited, effective strategies for the prevention of influenza particularly in infants under 6 months of age are needed to diminish the burden of disease in this age group.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.