Burden of illness in infants and young children hospitalized for respiratory syncytial virus: A rapid review.

Respiratory syncytial virus (RSV) infections are common among young children and represent a significant burden to patients, their families and the Canadian health system. Here we conduct a rapid review of the burden of RSV illness in children 24 months of age or younger. Four databases (Medline, Embase, Cochrane Database of Clinical Trials, ClinicalTrials.gov from 2014 to 2018), grey literature and reference lists were reviewed for studies on the following: children with or without a risk factor, without prophylaxis and with lab-confirmed RSV infection. Of 29 studies identified, 10 provided within-study comparisons and few examined clinical conditions besides prematurity. For infants of 33-36 weeks gestation (wGA) versus term infants, there was low-to-moderate certainty evidence for an increase in RSV-hospitalizations (n=599,535 infants; RR 2.05 [95% CI 1.89-2.22]; 1.3 more per 100 [1.1-1.5 more]) and hospital length of stay (n=7,597 infants; mean difference 1.00 day [95% CI 0.88-1.12]). There was low-to-moderate certainty evidence of little-to-no difference for infants born at 29-32 versus 33-36 wGA for hospitalization (n=12,812 infants; RR 1.20 [95% CI 0.92-1.56]). There was low certainty evidence of increased mechanical ventilation for hospitalized infants born at 29-32 versus 33-35 wGA (n=212 infants; RR 1.58, 95% CI 0.94-2.65). Among infants born at 32-35 wGA, hospitalization for RSV in infancy may be associated with increased wheeze and asthma-medication use across six-year follow-up (RR range 1.3-1.7). Children with versus without Down syndrome may have increased hospital length of stay (n=7,206 children; mean difference 3.00 days, 95% CI 1.95-4.05; low certainty). Evidence for other within-study comparisons was of very low certainty. In summary, prematurity is associated with greater risk for RSV-hospitalization and longer hospital length of stay, and Down syndrome may be associated with longer hospital stay for RSV. Respiratory syncytial virus-hospitalization in infancy may be associated with greater wheeze and asthma-medication use in early childhood. Lack of a comparison group was a major limitation for many studies.


Introduction
Respiratory syncytial virus (RSV) infections are common among young children (1,2), presenting as bronchiolitis, pneumonia, or other respiratory morbidity (3). Hospitalization due to RSV is a significant burden for patients, families and the Canadian health system (4).
Increased risk for RSV-hospitalization has been associated with age younger than one year (3,5), prematurity (6), chronic lung disease (7), congenital heart disease (8), other chronic conditions including cystic fibrosis, immunodeficiency (9)(10)(11)(12) and residence in Indigenous, northern or remote communities (13). These populations may also have higher rates of admission to intensive care units (ICU), requirements for respiratory support, and higher mortality attributable to RSV (9). RSV-hospitalization in the first two years of life has also been associated with wheezing in childhood (14- 16).
While no active vaccines exist for RSV prophylaxis, the monoclonal antibody palivizumab (Synagis ® , AstraZeneca) has demonstrated effectiveness in preventing RSV-hospitalization among some high risk populations (17,18). However, while efficacy of palivizumab (PVZ) in clinical trials appears to be high for children with some underlying clinical conditions, real-world evidence from observational studies is less certain (2), with wide variations in effectiveness. Due to the high numbers needed to treat in order to prevent hospitalization and the relatively high cost of PVZ, most jurisdictions use the intervention sparingly for select groups at highest risk of severe disease. Additionally, RSV vaccine development has been well under way, with some vaccine candidates undergoing phase 3 clinical trials (19). There is currently no global consensus on RSV risk groups and variable policies exist even within Canada.
The objective of this rapid review is to address the following question: What is the burden of RSV illness including long-term sequelae among children 24 months of age and younger without prophylaxis, and with or without risk factors for severe RSV disease, and for immunocompromised children younger than 18 years of age?
Findings from the review will help inform updated recommendations of Canada's National Advisory Committee on Immunization on the use of PVZ prophylaxis to prevent severe consequences of RSV infection. This evidence base will also be relevant for future deliberations on program design for anticipated RSV vaccines and newer monoclonal antibodies (19).

Methods
This review was guided by methods for reviews of interventions (20), overall prognosis (21), and risk of future event (prognosis) (22); a protocol was developed a priori (Supplement 1), and reviewed and approved by the National Advisory Committee on Immunization RSV Working Group. In light of the restricted literature search timeframe of interest to the review commissioners, we refer to the undertaken work as a rapid review.

Literature search
Searches were conducted on September 6, 2018, in Medline, Embase, Cochrane Database of Clinical Trials, ClinicalTrials. gov, and websites of international public health authorities (Supplement 2). Limits were applied for date of publication (January 1, 2014 to September 6, 2018) and language (English or French). The date limit was aimed at capturing outcomes just before and after significant changes in clinical practice stemming from the revised recommendations for PVZ prophylaxis by the American Academy of Pediatrics (23) as well as the Canadian Paediatric Society (2).

Study selection and eligibility criteria
Two reviewers independently screened titles and abstracts followed by full texts. Discrepancies were resolved by discussions.
Studies conducted in Organisation for Economic Co-operation and Development (OECD) countries, including observational studies and placebo groups of controlled trials were eligible for inclusion. Studies reporting on children 24 months of age and younger, with or without a risk factor of interest, or immunocompromised children 18 years of age and younger without PVZ prophylaxis and with lab-confirmed RSV infection were eligible. Children without RSV infection were eligible as a comparator group for long-term outcomes. Short-term outcomes included RSV-hospitalization, hospital length of stay, ICU admission and length of stay, oxygen support and duration, mechanical ventilation and duration, extracorporeal membrane oxygenation and duration, case fatality (death due to RSV), and complications from RSV infection (e.g. secondary infection). Long-term outcomes (minimum one-year follow-up) included self-reported, parent-reported or physician-diagnosed recurrent wheeze, atopic asthma, deterioration of pulmonary or cardiac function, and impaired growth or development. Detailed inclusion and exclusion criteria are in Supplement 3.

Data extraction, synthesis/analysis and risk of bias assessment
One reviewer extracted data with second-reviewer verification.
For dichotomous outcomes, we extracted the number of events and the number analysed in each eligible group, or relative measures (e.g. odds ratio) if crude events were not reported. For continuous outcomes, mean values for each time-point, and change scores, including standard deviations or measures of variability were extracted. Risk ratio (RR) with 95% confidence interval (CI) and mean difference (MD) were used for comparisons between groups.
Our primary interest was using data from studies that reported on two or more groups, either having different risk factors, or a risk group versus healthy term infants (within-study/direct comparisons). For similar comparisons reported by more than one study, data were pooled using the DerSimonian Laird random effects model inverse variance method with Mantel-Haenszel weighting. Risk differences were used when rare or zero events appeared in at least one study group. We also made comparisons between short-term outcomes in risk groups and healthy term infants reported by different studies (between-study/indirect comparisons). We used the double-arc sine transformation to pool single-group proportions across multiple studies. When no comparison was made, we report event proportions for the single group in these studies.
For outcomes where estimates were statistically significant, we calculated the absolute risk difference (24).

RAPID COMMUNICATION
Analyses were performed using Excel, Review Manager (version 5.3) and STATA (version 14.2).
Two reviewers independently assessed the risk of bias for each study, using a modified tool based on the Quality Assessment Tool for Observational Cohort and Cross-sectional Studies and the Quality In Prognosis Studies (QUIPS) tool (Supplement 4). Disagreements were resolved via consensus or third-reviewer consultation.

Certainty of evidence
Two reviewers independently assessed the certainty of evidence for each outcome (as high, moderate, low, or very low) from within-study comparisons (direct evidence), with disagreements resolved through consensus. The approach followed principles of the Grading of Recommendations Assessment, Development and Evaluation working group and considerations for a body of evidence that examines risk of future events (prognosis) (Supplement 5) (21).
We included three studies of children with RSV older than 24 months of age to capture evidence for immunocompromised populations: children three years old or younger with Down syndrome with or without known risk factors for RSV (50), and younger than 18 years old with liver transplantation (38) and sickle cell disease (49).
Six studies reported on long-term outcomes: healthy term infants with versus without RSV in infancy (27,28,31,52), premature infants with versus without RSV-hospitalization in infancy (32), and premature versus term infants hospitalized for RSV in their first RSV season (37).

Risk of bias
Risk of bias ratings are in Table 1 and Supplements 4 and 6.
Studies that reported incidence of RSV-hospitalization were at moderate-to-high risk of bias, mainly due to lack of blinding to childrens' risk status by healthcare providers that may have influenced admission to hospital. For other short-term outcomes, studies were mostly at moderate risk of bias (25,26,29,33,35,39,(45)(46)(47). Two studies were at high risk of bias due to concerns in more than one domain (41,44). Nearly all reported long-term outcomes (27,28,32,37,52) were at moderate risk of bias, arising from lack of blinding for patient or parent-reported outcomes and/or potential selection biases.
Short-term outcomes from within-study comparisons Table 2 summarizes evidence for short-term outcomes from within-study comparisons. Here we do not report further on findings having very low certainty of evidence.
Different degrees of prematurity: One study found little-to-no difference in RSV-hospitalization during their first RSV season for infants born at 29-32 compared with 33-36 weeks' gestation (wGA) (36). This study found little-to-no difference for hospital stay of less than one day versus one day or more between infants born at 29-32 or 33-36 wGA (36).
Another study of infants born at 29-32 versus 33-35 wGA who were hospitalized for RSV in the first year of life found littleto-no difference for ICU admission, but longer (although not statistically significant) ICU length of stay among the infants born at 29-32 wGA (26). There was a greater need for mechanical ventilation for hospitalized infants born at 29-32 wGA versus 33-35 wGA (26).
There were no studies of premature infants born before 29 wGA.
Premature versus term infants: One study found that being born late-premature (33-36 wGA) versus at term was associated with increased RSV-hospitalization in the first two years of life (42). The same study found slightly longer hospitalization for the preterm group (42).
Down syndrome: One study comparing children with Down syndrome without additional risk factors for RSV versus healthy children, all followed to three years of age, reported a higher hospitalization rate in those with Down syndrome (50). There was a discrepancy between the text and tables for RSV-hospitalization rates in these groups that could not be resolved due to unsuccessful attempts to contact the study authors (50). This study also found that RSV was associated with longer hospital length of stay among children with Down syndrome without other risk factors versus children without Down syndrome (50). For all cases of Down syndrome, including those with known risk factors for RSV, the authors conclude that Down syndrome is independently associated, after adjusting for known risk factors for RSV disease, with an increased risk for RSV-hospitalization (50). Of note, data on children younger than 24 months of age without risk factors were not isolated from those with additional risk factors, and therefore, were not used in our analysis.
Select short-term outcome comparisons: Other data Supplement 8 contains single risk group data and pooled analyses (when appropriate). Data for between-study comparisons are in Supplement 9. Findings for select outcomes are reported below; data on other short-term outcomes are in Supplement 8.
Single group proportions for RSV-hospitalizations were 5.1% in the first six months of life and 3.3% in the first two years of life for infants 29 to younger than 33 wGA (36,37) and 32/33 to 35

Complications
One study reported on complications associated with RSVhospitalization (Supplement 10).
Long-term outcome comparisons from within-study comparisons Tables 3 to 5 summarize evidence for long-term outcomes from within-study comparisons. No study reported on growth or impaired development. Here we do not report on findings having very low certainty evidence.      Prematurity: One study enrolled premature (32-35 wGA) infants with or without hospitalization for RSV infection before 12 months of age to examine several long-term outcomes (32). Data were collected through telephone calls every four months and annual visits until the 6 th year of life. The authors analysed data both across the five years and only within the 6 th year. All findings offered low certainty evidence.

RAPID COMMUNICATION
Across years two through six, associations were found between RSV-hospitalization and increased risk for parent or physician-reported simple wheeze, recurrent wheeze, severe wheeze and any/all wheeze (32). When examining the 6 th year only, there was little-to-no difference in parent or physician-reported simple wheeze, recurrent wheeze, severe wheeze and any/all wheeze (32). This study also compared groups for parent-reported asthma-associated medication use across years two through six. There were associations with increased risk from RSV-hospitalization for use of bronchodilators, inhaled corticosteroids, oral corticosteroids and leukotriene antagonists (32). Through lung function testing using spirometry, there was little-to-no difference in severe respiratory morbidity at six years of age between groups (32).
RSV without risk factors: Pooled data from two studies of children with versus without RSV-hospitalization at younger than 24 months of age found low certainty of an increase in self-reported asthma in adulthood (27,28). Of note, there was no difference in physician-diagnosed asthma (considered more reliable than patient-reported asthma) (28), but certainty of evidence was very low for this outcome. An association was also found between RSV and lower pre-bronchodilation mean percent of predicted forced expiratory volume in one second (FEV 1 ), but there was little-to-no difference in change in FEV 1 from pre to post-bronchodilation (27,28). The RSV was associated with lower predicted forced vital capacity, but not fractional exhaled nitric oxide (27,28).
Long-term outcome comparisons: Other data Supplement 11 contains single group data and pooled analyses (when appropriate). We did not conduct analyses for between-study comparator groups, since single studies that contributed to each long-term outcome were represented among within-study comparisons.

Summary of findings and limitations
Few studies contributed data for within-study comparisons of outcomes of interest. There was moderate-to-low certainty that RSV is associated with a small increase in hospitalization and length of stay among moderate-to-late preterm (33-36 wGA) compared with term infants. There was moderate-to-low certainty evidence for no significant differences in hospitalizations between infants born at 29-32 wGA versus 33-36 wGA. Low certainty of evidence was found for a slight increase in mechanical ventilation among those born at 29-32 wGA versus 33-35 wGA and hospitalized for RSV prior to 12 months of age. Low certainty evidence was found for increased hospital length of stay among children younger than three years of age with versus without Down syndrome. There was low certainty evidence for increased wheezing and asthma medication use from two to six years of age among RSV-hospitalized versus non-hospitalized premature (32-35 wGA) infants, although there was little-to-no difference for these outcomes in the 6 th year of follow-up. Low certainty evidence was found for decreased lung function measurements before bronchodilation but changes in measurements after bronchodilation did not differ between groups. Very low certainty evidence was found for other long-term outcomes comparing different risk groups.
Single studies contributed data for most outcomes, where populations with rare conditions (e.g. cystic fibrosis) often represent small/under-powered sample sizes, precluding investigation of heterogeneity among studies for important population and RSV characteristics, or consistency in findings.
The paucity of studies on clinical conditions other than prematurity is a limitation of the evidence base. We also did not find studies of premature infants born before 29-30 weeks gestation, or of children with chronic lung disease of prematurity or congenital heart disease, groups for whom prophylaxis is now recommended in the United States and in Canada (2,23).
Retrospective study designs utilizing older data (i.e. pre-2014) were included, and may reflect different practices (e.g. prophylaxis, RSV-testing, standard of care) over time and across countries and settings. Detection of RSV infection may be impacted by variation in testing methods, including types of tests and indications for testing, and seasonal and geographic variability. Among tested individuals, the proportion of patients with viral or bacterial co-infections may be an important confounder in etiology of outcome severity. Lack of blinding of healthcare providers to risk status may influence rates of hospitalization and possibly other care parameters, particularly among children with known RSV risk factors.

Comparison with other reviews
A series of systematic reviews of publications from 1995 to 2015 found that RSV-hospitalization is associated with significant morbidity among children younger than 18 years old in Western countries (Canada, United States, Europe), particularly for young children with prematurity, chronic lung disease of prematurity and congenital heart disease (6)(7)(8)(9). Whereas the current work focused on children younger than two years of age with a single risk factor, these reviews also included studies of children up to 18 years of age. Our review scope searched comparatively more recent publications (2014-2018) and covered a broader geographic area by including high-income (OECD) countries.

Future research
Based on current evidence, there is a need for studies to focus on the burden of RSV disease among children with underlying chronic conditions, for some of which data on risk are contradictory or non-existing. Assessments of current RSV surveillance activities in Canada have identified data gaps for particular populations, including children with underlying medical conditions and those living in Indigenous, northern or remote communities (19). Gaps will need to be filled in preparation for monitoring of RSV vaccine effectiveness in the future.

Conclusion
Prematurity is associated with increased risk for RSVhospitalization in infancy and increased hospital length of stay, and may be associated with increased wheeze and asthmamedication use at up to six years of age. Down syndrome may be associated with longer hospital length of stay.

Supplemental material
These documents can be accessed on the Supplemental material file.