CONTEXT:

Respiratory virus (RV) detection tests are commonly used in hospitalized children to diagnose viral acute respiratory infection (ARI), but their clinical utility is uncertain.

OBJECTIVES:

To systematically review and meta-analyze the impact of RV test results on antibiotic consumption, ancillary testing, hospital length of stay, and antiviral use in children hospitalized with severe ARI.

DATA SOURCES:

Seven medical literature databases from 1985 through January 2018 were analyzed.

STUDY SELECTION:

Studies in children <18 years old hospitalized for severe ARI in which the clinical impact of a positive versus negative RV test result or RV testing versus no testing are compared.

DATA EXTRACTION:

Two reviewers independently screened titles, abstracts, and full texts; extracted data; and assessed study quality.

RESULTS:

We included 23 studies. High heterogeneity did not permit an overall meta-analysis. Subgroup analyses by age, RV test type, and viral target showed no difference in the proportion of patients receiving antibiotics between those with positive versus negative test results. Stratification by study design revealed that RV testing decreased antibiotic use in prospective cohort studies (odds ratio = 0.58; 95% confidence interval: 0.45–0.75). Pooled results revealed no conclusive impact on chest radiograph use (odds ratio = 0.71; 95% confidence interval: 0.48–1.04). Results of most studies found that positive RV test results did not impact median hospital length of stay, but they may decrease antibiotic duration. Nineteen (83%) studies were at serious risk of bias.

LIMITATIONS:

Low-quality studies and high clinical and statistical heterogeneity were among the limitations.

CONCLUSIONS:

Higher-quality prospective studies are needed to determine the impact of RV testing on antibiotic use in children hospitalized with severe ARI.

Acute respiratory infections (ARIs) are a leading cause of global childhood morbidity. Severe ARIs, most notably acute lower respiratory infections (ALRIs), are 1 of the most common reasons for hospitalization and mortality in children.13  The Global Burden of Disease Study estimated 29.2 million cases of ALRIs in 2015. This resulted in 2.7 million deaths, of which >700 000 occurred in children <5 years of age.4 

Although bacteria can cause severe ARIs, ARIs are most often caused by viruses.5  Pneumonia, which can be viral or bacterial in etiology, and acute viral bronchiolitis account for most of the severe ARI global burden of disease in young children.5,6  Globally, respiratory syncytial virus (RSV) is the most common etiology of severe childhood ARI.79 

A major challenge for clinicians is to distinguish viral from bacterial causes of severe ARI because their presentations overlap.5  This, in addition to the risk of bacterial superinfection in severe viral ARI, leads to the frequent empirical use of antibiotics in children who have only a viral infection.1012  Concerns regarding possible bacterial etiologies are also associated with additional, often unnecessary, ancillary tests, prolonged duration of hospitalization, and increased health care costs.13,14 

Antibiotic prescribing in severe viral ARI is a key contributor to the major public health issue of antibiotic overuse and increasing bacterial resistance. Antibiotic overuse also exposes patients to potential harms, such as toxicity and adverse reactions.15,16  The hospital setting is of particular importance because resistant pathogens are transmitted to vulnerable patients.

Respiratory virus (RV) detection tests are commonly used in children hospitalized with severe ARI.17,18  However, current literature regarding the impact of RV tests on use of antibiotics and/or ancillary tests reveals discordant results.13,1822  It is imperative to evaluate the impact of such tests to inform evidence-based clinical practice guidelines for pediatric respiratory infections, which currently offer vague or conflicting recommendations on their use.23,24  A reduction in antibiotic use and/or ancillary testing would support the routine use of RV tests because they could help curb antibiotic overuse and reduce costly ancillary testing. However, if such tests do not improve the management of patients with severe ARI, their routine use may not be justified.

We hypothesized that RV testing decreases unnecessary antibiotic use in hospitalized pediatric patients with severe ARI. Thus, we conducted a systematic review and meta-analysis to determine the impact of RV testing on antibiotic consumption, ancillary testing, length of hospital stay, and influenza antiviral prescribing in children hospitalized with severe ARI.

The protocol was developed according to the Preferred Reporting Items for Systematic Review and Meta-analysis Protocols (PRISMA-P) statement and registered with the international prospective register of systematic reviews (PROSPERO; registration number: CRD42018088273).

We searched Ovid Medline, Ovid Embase, the Cochrane Central Register of Controlled Trials, Web of Science, BIOSIS Previews, Scopus, Clinicaltrials.gov, and the International Clinical Trials Registry Platform from 1985 to January 8, 2018. We developed the search strategy (Supplemental Information) in collaboration with a health sciences librarian. Lastly, we used Scopus and Google Scholar for forward citation searching.

Study Design and Participants

We included original published abstracts and full reports that evaluated the impact of respiratory viral testing in hospitalized children (<18 years of age) with ARI (defined as an illness of <7 days’ duration with respiratory symptoms suggestive of infection) on the following outcomes: antibiotic use, ancillary testing, hospital length of stay, and/or antiviral use. Because severe ARI requiring hospitalization is almost always due to ALRI, we included patients tested for any ARI because many studies did not specify ARI type (ALRI versus upper respiratory infection). Eligible study designs included randomized controlled trials, quasi-randomized controlled trials, prospective and retrospective cohort studies, case-control studies, and cross-sectional studies. We excluded studies of patient populations restricted to specific underlying comorbidities that would increase the likelihood of receiving antibiotics, such as immunosuppression or cystic fibrosis.

Exposures and Outcomes

Eligible studies assessed the impact of exposure to results of RV testing (positive versus negative result) or administration of RV tests (yes versus no). RV tests could include shell vial cultures, immunofluorescence assays, nucleic acid amplification tests (NAATs), and rapid antigen detection tests (RADTs). Our primary outcome was antibiotic use, defined in 3 ways: (1) the proportion of patients prescribed antibiotics, (2) the duration of antibiotic use in days, and (3) the proportion of patients in whom empirical antibiotics were stopped when a virus was detected. Our secondary outcomes were use of ancillary diagnostic tests (proportion), hospital length of stay (days), and use of antiviral agents (proportion).

Two reviewers independently screened titles and abstracts (first screen) and full-text reports (second screen). Discrepancies were resolved by consensus or by an arbitrator.

Two reviewers piloted the data extraction form with 10% of included studies. The form was then modified and finalized. These reviewers independently extracted the data on study design, exposures, population characteristics, setting, sample size, author, publication year, characteristics of RV test (turnaround time and cost), study quality, outcomes, and funding. We categorized the age of the study population as ≤1 year old if >75% of patients in the sample were ≤1 year old. We attempted to identify the timing of antibiotic use and/or ancillary testing (before versus after RV test results or performance of RV testing) using data reported in published articles. When no information was available, we contacted authors to retrieve missing data.

Two reviewers independently assessed the quality of cohort studies using the Cochrane risk of bias in nonrandomized studies of interventions tool.25  Items used to assess quality include biases related to confounding (no adjustment for age, severity of illness, or presence of comorbidities), patient selection, classification of interventions, deviations from intended interventions, missing data, outcome measurement, and selective reporting. The reviewers judged the risk of bias for each study as “low risk,” “moderate risk,” “serious risk,” “critical risk of bias,” or “no information.” Cohort and case-control studies were also assessed by using the Newcastle-Ottawa scale, which focuses on the selection and comparability of study groups on outcome ascertainment. The highest-quality studies were awarded up to 9 stars.26 

The impacts of the 2 exposure groups (RV test results and performance of RV tests) were analyzed and presented separately. Associations of the exposure with dichotomous outcomes were expressed as odds ratios (ORs) with 95% confidence intervals (CIs), and continuous outcomes were expressed as standard mean differences with 95% CIs. We pooled studies that were clinically homogeneous in terms of patient populations, diagnosis type, exposure type, and outcomes and assessed statistical heterogeneity using the I2 statistic. If sufficient studies were available (≥3 studies), we performed meta-analyses using a random-effects model with the restricted maximum likelihood method to estimate between-study heterogeneity and reported the overall effect of each exposure as ORs with 95% CIs. We conducted meta-analyses within prespecified strata defined by patient age (patients ≤1 year old were considered infants, and patients >1 year old were considered children), type of RV test (shell vial cultures, immunofluorescence assays, NAATs, and RADTs), RV test target (influenza A and/or B, RSV, and >1 viral target), study design (prospective versus retrospective cohort study), location of RV test (point-of-care test, defined as testing done outside the laboratory by nonlaboratory personnel and in close proximity to patient care27  versus laboratory-based test), turnaround time (≤6 vs >6 hours), and source of funding (industry versus other). Publication bias was assessed by using funnel plots and the Egger test. We conducted a sensitivity analysis restricted to studies deemed to be at low risk of bias. Narrative summaries were presented for results that could not be meta-analyzed. A post hoc meta-analysis of studies exclusively evaluating patients with bronchiolitis was performed. Analyses were conducted by using R version 3.4.3 and the metafor package.28 

We identified 7697 records (Supplemental Figs 5–11), of which 23 articles met the inclusion criteria: 9 prospective cohort studies, 12 retrospective cohort studies, 1 mix of retrospective and prospective cohort designs, and 1 case-control study (Table 1).21,2950 

TABLE 1

Characteristics of Included Studies

StudyCountrySample SizeAge, mo, Median (IQR) or Mean (SD)Primary Diagnosis, %Percent With ComorbiditiesIntCompViral Test TypeViral Test Turnaround Time, hOutcomes Reporteda
Prospective cohort studies           
 Adcock et al29  United States 160 Int: 3.1 (NR); Comp: 3.6 (NR) NR RSV+ RSV− RADT 1, 2, 5 
 Bozdemir et al31  Turkey 671 Int: 4 (NR); Comp: 7 (NR) Bronchiolitis (48); pneumonia (42) NR RSV+ RSV− RADT NR 
 Bueno et al32  Brazil 21 Int: 2.4 (2.5)b; Comp: 3.2 (1.9)b NR 43.0 RSV+ RSV− NAAT NR 1, 5 
 Hatipoğlu et al36  Turkey 147 Int: 13.6 (15.9)c; Comp: 14.7 (15.8)c Bronchiolitis (56); pneumonia (44) 17.0 Virus+ Virus− RADT, immunofluorescence assay, NAAT, culture NR 1, 5 
 Manji et al37  United States 603 NR NR NR RSV+ RSV− Immunofluorescence assay, culture NR 1, 2, 5 
 Nitsch-Osuch et al40  Poland 52 NR Influenza (29); pneumonia (21) 12.7 lnfluenza+ Influenza− NAAT NR 1, 6 
 Suntarattiwong et al45 d Thailand 354 First Int: 7.3 (2.8)c; second Int: 7 (3.4)c; Comp: 7 (3)c Pneumonia (73) 23.2 RSV+ and/or influenza+ RSV− and/or influenza− NAAT NR 1, 5, 6 
 Tsung et al48  Hong Kong 469 NR Bronchiolitis (69); pneumonia (10) NR Virus+ Virus− Immunofluorescence assay 24 
 van de Pol et al49  Netherlands 38 Total: 1.6 (8.8)b Bronchiolitis (53); pneumonia (48) 36.6 Virus+ Virus− Immunofluorescence assay, NAAT 
Retrospective cohort studies           
 Azzarone et al30  United States 462 Total: 6.5 (NR)c Bronchiolitis (100) RSV+ RSV− RADT NR 1, 4 
 Byington et al33  (season 1) United States 229 Total: 29 (NR)c Bronchiolitis (37); pneumonia (12) NR Virus+ Virus− Immunofluorescence assay 8.6 
 Ferronato et al34  Brazil 211 Total: 4 (2.7)c Bronchiolitis (100) RSV+ Virus− Immunofluorescence assay 40 1, 3 
 Flaherman et al35  United States 1727 NR Bronchiolitis (100) 10.3 RSV+ RSV− Immunofluorescence assay NR 
 Milić et al39  Croatia 193 Int: 2.3 (1.4–3.2); Comp: 4.8 (2.1–7.5) Bronchiolitis (100) RSV+ RSV− RADT NR 
 McCulloh et al38  (test versus no test) United States 1727 Int: 3.3e; Comp: 4.1e Bronchiolitis (28); pneumonia (50) 43 RV test No RV test NAAT 20 1, 4, 6 
 McCulloh et al38  (positive versus negative result) United States 809 Total: 3.3e Bronchiolitis (36); pneumonia (55) 44.8 Virus+ Virus− NAAT 20 1, 4 
 Paul et al41 f United Kingdom 319 First Int: 2.9 (NR); second Int: 3.1 (NR); Comp: 3.6 (NR) Bronchiolitis (100) 21.0g RSV+ and/or rhinovirus+ Virus− RADT, PCR NR 1, 4, 5 
 Schulert et al42  United States 717 Int: 12 (NR); Comp: 36 (NR) Bronchiolitis and RSV pneumonia (14.6); other pneumonia (17) 12.1 Virus+ Virus− NAAT 24 1, 2, 5 
 Schulert et al43  United States 202 Int: 12 (3.6–48); Comp: 36 (3.6–96) Pneumonia (100) Virus+ Virus− NAAT 24 1, 2, 5 
 Thibeault et al21  Canada 198 NR Bronchiolitis (91) 39.0 RSV+ RSV− RADT, NAAT NR 
 Tresoldi et al46  Brazil 61 Int: 10.6 (NR); Comp: 8.8 (NR) NR 54.0g lnfluenza+ Influenza− NAAT NR 
 Walls et al50  New Zealand 237 Total: 26 (NR)c Pneumonia (100) RV test No RV test NAAT NR 1, 3 
Case-control studies           
 Sulieman et al44  United States 200 NR NR NR RSV+ and/or influenza+ RSV− and/or influenza− Culture 48 
Mix of prospective and retrospective           
 Tsolia et al47  Greece 473 Int: 2.8 (NR); Comp: 4.5 (NR) Bronchiolitis (100) 2.5 RSV+ RSV− Immunofluorescence assay NR 1, 4, 5 
StudyCountrySample SizeAge, mo, Median (IQR) or Mean (SD)Primary Diagnosis, %Percent With ComorbiditiesIntCompViral Test TypeViral Test Turnaround Time, hOutcomes Reporteda
Prospective cohort studies           
 Adcock et al29  United States 160 Int: 3.1 (NR); Comp: 3.6 (NR) NR RSV+ RSV− RADT 1, 2, 5 
 Bozdemir et al31  Turkey 671 Int: 4 (NR); Comp: 7 (NR) Bronchiolitis (48); pneumonia (42) NR RSV+ RSV− RADT NR 
 Bueno et al32  Brazil 21 Int: 2.4 (2.5)b; Comp: 3.2 (1.9)b NR 43.0 RSV+ RSV− NAAT NR 1, 5 
 Hatipoğlu et al36  Turkey 147 Int: 13.6 (15.9)c; Comp: 14.7 (15.8)c Bronchiolitis (56); pneumonia (44) 17.0 Virus+ Virus− RADT, immunofluorescence assay, NAAT, culture NR 1, 5 
 Manji et al37  United States 603 NR NR NR RSV+ RSV− Immunofluorescence assay, culture NR 1, 2, 5 
 Nitsch-Osuch et al40  Poland 52 NR Influenza (29); pneumonia (21) 12.7 lnfluenza+ Influenza− NAAT NR 1, 6 
 Suntarattiwong et al45 d Thailand 354 First Int: 7.3 (2.8)c; second Int: 7 (3.4)c; Comp: 7 (3)c Pneumonia (73) 23.2 RSV+ and/or influenza+ RSV− and/or influenza− NAAT NR 1, 5, 6 
 Tsung et al48  Hong Kong 469 NR Bronchiolitis (69); pneumonia (10) NR Virus+ Virus− Immunofluorescence assay 24 
 van de Pol et al49  Netherlands 38 Total: 1.6 (8.8)b Bronchiolitis (53); pneumonia (48) 36.6 Virus+ Virus− Immunofluorescence assay, NAAT 
Retrospective cohort studies           
 Azzarone et al30  United States 462 Total: 6.5 (NR)c Bronchiolitis (100) RSV+ RSV− RADT NR 1, 4 
 Byington et al33  (season 1) United States 229 Total: 29 (NR)c Bronchiolitis (37); pneumonia (12) NR Virus+ Virus− Immunofluorescence assay 8.6 
 Ferronato et al34  Brazil 211 Total: 4 (2.7)c Bronchiolitis (100) RSV+ Virus− Immunofluorescence assay 40 1, 3 
 Flaherman et al35  United States 1727 NR Bronchiolitis (100) 10.3 RSV+ RSV− Immunofluorescence assay NR 
 Milić et al39  Croatia 193 Int: 2.3 (1.4–3.2); Comp: 4.8 (2.1–7.5) Bronchiolitis (100) RSV+ RSV− RADT NR 
 McCulloh et al38  (test versus no test) United States 1727 Int: 3.3e; Comp: 4.1e Bronchiolitis (28); pneumonia (50) 43 RV test No RV test NAAT 20 1, 4, 6 
 McCulloh et al38  (positive versus negative result) United States 809 Total: 3.3e Bronchiolitis (36); pneumonia (55) 44.8 Virus+ Virus− NAAT 20 1, 4 
 Paul et al41 f United Kingdom 319 First Int: 2.9 (NR); second Int: 3.1 (NR); Comp: 3.6 (NR) Bronchiolitis (100) 21.0g RSV+ and/or rhinovirus+ Virus− RADT, PCR NR 1, 4, 5 
 Schulert et al42  United States 717 Int: 12 (NR); Comp: 36 (NR) Bronchiolitis and RSV pneumonia (14.6); other pneumonia (17) 12.1 Virus+ Virus− NAAT 24 1, 2, 5 
 Schulert et al43  United States 202 Int: 12 (3.6–48); Comp: 36 (3.6–96) Pneumonia (100) Virus+ Virus− NAAT 24 1, 2, 5 
 Thibeault et al21  Canada 198 NR Bronchiolitis (91) 39.0 RSV+ RSV− RADT, NAAT NR 
 Tresoldi et al46  Brazil 61 Int: 10.6 (NR); Comp: 8.8 (NR) NR 54.0g lnfluenza+ Influenza− NAAT NR 
 Walls et al50  New Zealand 237 Total: 26 (NR)c Pneumonia (100) RV test No RV test NAAT NR 1, 3 
Case-control studies           
 Sulieman et al44  United States 200 NR NR NR RSV+ and/or influenza+ RSV− and/or influenza− Culture 48 
Mix of prospective and retrospective           
 Tsolia et al47  Greece 473 Int: 2.8 (NR); Comp: 4.5 (NR) Bronchiolitis (100) 2.5 RSV+ RSV− Immunofluorescence assay NR 1, 4, 5 

Comp, comparison; Int, intervention; IQR, interquartile range; NR, not reported; PCR, polymerase chain reaction; +, positive; –, negative.

a

1, proportion who received antibiotics; 2, duration of antibiotics; 3, proportion in whom antibiotics were stopped; 4, proportion who received ancillary tests; 5, length of stay; and 6, proportion who received antiviral agents.

b

Median (IQR) reported.

c

Mean (SD) reported.

d

First intervention = intervention group with influenza, and second intervention = intervention group with RSV.

e

Study authors do not state if mean or median. It includes prematurity because the study authors do not present underlying conditions (chronic pulmonary disease, congenital heart disease, etc) separately.

f

First intervention = intervention group with RSV, and second intervention = intervention group with rhinovirus.

g

Twenty-one percent of patients in control group had ≥2 risk factors (defined as prematurity, congenital heart disease, chronic lung disease, and/or genetic conditions).

In 22 studies (96%), authors assessed the impact of a positive versus a negative viral test result, of which 2 (8%) were published abstracts. In 1 of these studies both exposures of interest were assessed.38  The median age of participants ranged from 2 to 12 months for positive RV test result groups and 3.6 to 36 months for negative RV test result groups. In 6 studies (26%) the age of the participant was not reported. The proportion of patients with comorbidities, ranging from 0% to 54%, was reported in 18 studies (78%). RV tests evaluated included RADTs (4 studies; 17%), NAATs (6 studies; 26%) and immunofluorescence assays (6 studies; 26%). Turnaround time, ranging from 2 to 48 hours, was reported in 10 studies (43%). The proportion of participants who received antibiotics was reported in 22 studies (96%).

The impact of RV testing compared with no testing was assessed in 2 studies (9%).38,50  NAATs and reported turnaround times (7 and 20 hours) and the proportion of participants who received antibiotics were used in both studies.

Most studies presented overall moderate to serious risk of bias, primarily because of confounding and selection of reported results (Supplemental Figs 5–11). None of the studies were at critical risk of bias, and only 1 study was at overall low risk of bias. The median score for the Newcastle-Ottawa scale for cohort studies was 7 (interquartile range 7–7).

The impact of RV test results on the proportion of patients prescribed antibiotics was reported in 20 studies (87%). Seven studies (30%) revealed that a positive RV test result significantly decreased the odds of receiving antibiotics, ranging from an OR of 0.25 (95% CI: 0.07–0.92) to an OR of 0.65 (95% CI: 0.54–0.79; Fig 1).29,3538,40,41  Increased odds of receiving antibiotics was found in 2 studies (9%), ranging from an OR of 3.33 (95% CI: 1.07–10.34) to an OR of 6.82 (95% CI: 1.68–27.66).43,46  No impact on antibiotic prescription rates was reported in 11 studies (48%).3033,39,42,44,45,4749  Because of high levels of clinical and statistical heterogeneity, we did not perform a meta-analysis of all included studies. Furthermore, we could not perform a sensitivity analysis of high-quality studies because only 1 was judged to be at low risk of bias.

FIGURE 1

Results of individual studies on the proportion of patients receiving antibiotics among those with a positive versus negative RV test result. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

FIGURE 1

Results of individual studies on the proportion of patients receiving antibiotics among those with a positive versus negative RV test result. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

Subgroup analyses revealed no difference in frequency of antibiotic prescription when data were stratified by age (Fig 2), type of RV test (Supplemental Figs 5–11), and viral target (Supplemental Figs 5–11). Stratifying by study design type (Fig 3), the pooled OR for antibiotic use was 0.58 (95% CI: 0.45–0.75; I2 = 25%) for prospective cohort studies (9 studies) and 1.12 (95% CI: 0.70–1.80; I2 = 82%) for retrospective cohort studies (9 studies). We did not pool subgroups of studies that used multiple RV tests or that targeted influenza specifically because the types of RV test combinations differed and their number was not sufficient (2 studies), respectively. Two studies were not included in the subgroup analysis by study design because they used either a case-control design or a mix of prospective and retrospective cohort designs.44,47 

FIGURE 2

Subgroup analysis by age: proportion of patients receiving antibiotics among those with a positive versus negative RV test result. Tests are for subgroup differences: P = .96. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

FIGURE 2

Subgroup analysis by age: proportion of patients receiving antibiotics among those with a positive versus negative RV test result. Tests are for subgroup differences: P = .96. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

FIGURE 3

Subgroup analysis by study design: proportion of patients receiving antibiotics among those with a positive versus negative RV test result. Tests are for subgroup differences: P = .02. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

FIGURE 3

Subgroup analysis by study design: proportion of patients receiving antibiotics among those with a positive versus negative RV test result. Tests are for subgroup differences: P = .02. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

Finally, we could not perform a subgroup analysis of point-of-care testing, source of funding (insufficient data reported), and turnaround time (<3 studies per stratum). Turnaround times of >6 hours were reported in 6 studies (26%). Of these, McCulloh et al38  found that a positive RV test result significantly decreased the odds of receiving antibiotics (OR = 0.53; 95% CI: 0.34–0.81), Schulert et al43  reported an increase in odds (OR = 3.33; 95% CI: 1.07–10.34), and 4 studies revealed no impact.33,42,44,48  Turnaround times of ≤6 hours were reported in 2 studies, with ORs ranging from 0.42 (95% CI: 0.22–0.80)29  to 0.52 (95% CI: 0.14–1.92).49 

The timing of antibiotic use in relation to RV test results was reported in 5 studies. Decreased odds of receiving empirical antibiotics before test results were available (OR = 0.42; 95% CI: 0.22–0.80)29  or no impact31,39  was reported in 3 studies. Of these, 2 were prospective cohort studies,29,31  and 1 was a retrospective cohort study.39  In 2 prospective cohort studies, the proportion of patients who received antibiotics after RV test results was reported.48,49  Both found nonsignificant point estimates favoring decreased antibiotic prescription after a positive test result, ranging from an OR of 0.52 (95% CI: 0.14–1.92) to an OR of 0.62 (95% CI: 0.38–1.03).

Lastly, we performed post hoc subgroup analyses of studies by type of severe ARI patient population (bronchiolitis or pneumonia) and of studies that tested for influenza viruses. In five studies only patients with bronchiolitis were examined (Fig 4), and the pooled OR was 0.96 (95% CI: 0.59–1.56; I2 = 84%). Schulert et al43  included only patients with pneumonia and found that a positive RV test result increased the odds of antibiotic prescription (OR = 3.33; 95% CI: 1.07–10.34). In 2 studies, researchers reported on testing performed exclusively for influenza viruses. Nitsch-Osuch et al40  reported decreased odds of antibiotic prescription (OR = 0.25; 95% CI: 0.07–0.92), whereas Tresoldi et al46  reported increased odds (OR = 6.82; 95% CI: 1.68–27.66). Influenza was tested for, in addition to other viruses, in 12 studies (52%). After stratifying by study design type (Supplemental Figs 5–11), the pooled OR for antibiotic use was 0.54 (95% CI: 0.37–0.80; I2 = 44%) for prospective cohort studies (6 studies) and 1.21 (95% CI: 0.58–2.52; I2 = 84%) for retrospective.

FIGURE 4

Subgroup analysis of studies examining only patients with bronchiolitis. Proportion of patients receiving antibiotics among those with a positive versus negative RV test result. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

FIGURE 4

Subgroup analysis of studies examining only patients with bronchiolitis. Proportion of patients receiving antibiotics among those with a positive versus negative RV test result. ATB, antibiotic virus–, virus-negative; virus+, virus-positive.

Administration of an RV test to no testing was compared in 2 studies, and the proportion of patients prescribed antibiotics was reported for both. Whereas the study of McCulloh et al38  found that testing increased antibiotic prescriptions (OR = 1.36; 95% CI: 1.13–1.64), Walls et al50  reported the opposite in patients with pneumonia (OR = 0.21; 95% CI: 0.09–0.48).

The impact of RV test results on the duration of antibiotic use was reported in 5 studies (22%), but they could not be meta-analyzed because duration was presented as a median with no measure of variability. Four studies found a statistically significant decrease in median durations between groups with a positive versus negative RV test result that ranged from 1 to 3.1 days,29,33,37,42  whereas the study of Schulert et al43  found no difference between groups (2.6 vs 2.7 days; P = 1.0).

Neither study comparing RV testing to no testing reported the duration of antibiotic use.

The impact of RV test results on the proportion of patients in whom antibiotics were stopped was reported in 2 studies (9%). Whereas the study of Ferronato et al34  found that a positive RV test result increased the odds of stopping antibiotics (OR = 7.83; 95% CI: 2.89–21.25), Thibeault et al21  reported no difference (OR = 0.69; 95% CI: 0.29–1.64).

Among the studies whose exposure was testing versus no testing, Walls et al50  showed a significant increase in the odds of stopping antibiotics in tested patients (OR = 5.28; 95% CI: 1.49–18.68).

The impact of RV test results on the proportion of patients for whom chest radiographs were ordered was reported in 3 studies (13%). Pooled results (Supplemental Figs 5–11) did not show a significant effect (OR = 0.71; 95% CI: 0.48–1.04; I2 = 52%).

McCulloh et al38  found that administering RV testing was associated with ordering more ancillary tests, including chest radiographs (OR = 2.30; 95% CI: 1.85–2.85), blood tests (OR = 2.53; 95% CI: 2.08–3.08), urine cultures (OR = 2.03; 95% CI: 1.62–2.53), and throat cultures (OR = 3.93; 95% CI: 2.33–6.63).

The impact of RV test results on hospital length of stay was reported in 11 studies (48%), but no meta-analysis was performed because only 2 studies presented the results as means. A difference in mean or median length of stay between exposure groups was not found in 9 studies. Manji et al37  found a reduced median duration of hospitalization among patients with a positive versus negative RV test result (2 vs 3 days; P = .035). Tsolia et al47  found a prolonged mean length of stay for patients with a positive RV test result (standard mean difference = 0.23; 95% CI: 0.04–0.41).

No studies were reported comparing hospital length of stay between patients receiving an RV test and no test.

The impact of RV test results on the proportion of patients who received influenza antiviral agents was reported in 2 studies (9%). Whereas Nitsch-Osuch et al40  found that no patients received antiviral agents, Suntarattiwong et al45  found no significant difference between exposure groups (OR = 2.95; 95% CI: 0.65–12.54).

McCulloh et al38  reported the impact of RV testing on antiviral use, and found significantly increased odds of receiving antiviral agents in patients tested (OR= 10.47; 95% CI: 7.13–15.39).

The funnel plot for the proportion of patients who received antibiotics did not reach significance for asymmetry (P = .41; Supplemental Figs 5–11).

Our systematic review identified 22 studies in which the impact of a positive versus negative RV test result was evaluated and 2 studies in which the impact of administering an RV test or not was evaluated. Overall, our results did not reveal that a positive RV test result is associated with a reduction in antibiotic prescription for hospitalized pediatric patients with severe ARI. In addition, we found through our meta-analysis that a positive RV test result had no impact on chest radiograph use. Nevertheless, researchers in individual studies indicated that a positive RV test result may decrease the duration of antibiotic use and increase the likelihood of stopping antibiotics. However, RV test results do not seem to affect hospital length of stay. We could not determine the impact of administering an RV test or not on antibiotic use, ancillary testing, and length of stay because of an insufficient number of studies identified.

The subgroup meta-analyses revealed no difference in the proportion of patients receiving antibiotics between those with positive versus negative test results except when we stratified by study design. We speculate that the observed association between a positive RV test result and a decrease in antibiotic use among prospective cohort studies may be in part because, in these studies, it was possible to observe the proportion of patients for whom antibiotics were prescribed and ancillary testing was ordered after the release of RV test results. Conversely, it is unlikely that authors of most of the retrospective studies could ascertain whether antibiotics were prescribed before RV testing or after results were received.

Furthermore, because none of the studies were randomized, it is possible that any effect of RV testing on the use of antibiotics, ancillary testing, and length of stay may have been masked by the presence of other variables associated with RV testing and antibiotic use (eg, comorbidities, ARI type, and severity of illness). Fourteen studies (61%) in our review included patients with comorbidities such as cancer, hematologic conditions, and chronic pulmonary disease. Because of their compromised immune systems and clinical frailty, these patients are more likely to receive antibiotics, have ancillary tests performed, and need longer hospital admissions regardless of RV test results. Schulert et al42  observed that a positive RV test result was associated with a longer length of stay for hematology and/or oncology patients compared with those in general wards (P = .03). Moreover, certain types of severe ARI (eg, pneumonia) are more likely to have a mixed viral-bacterial source. This increases the likelihood of empirical antibiotics and ancillary testing to rule out a concomitant or secondary bacterial infection. ARI disease severity can also increase antibiotic use and prolong length of stay. Bradshaw et al17  found through a survey that for children with bronchiolitis admitted to Canadian PICUs, 36% of intensivists would prescribe antibiotics for moderately ill patients compared with 71% for intubated patients. Unfortunately, we were unable to perform a subgroup analysis to control for these possible confounders because we did not have access to the primary data. Lastly, cohort studies may have been subject to confounding by indication, in which patients for whom doctors had a strong suspicion of bacterial infection were both more likely to be tested and to receive antibiotics and/or ancillary testing.

Most severe ARI diagnoses in the included studies were bronchiolitis and pneumonia. Current American Academy of Pediatrics guidelines specifically recommend against routine RV testing for children with bronchiolitis because “knowledge gained from such testing rarely alters management decisions.”51  We support this statement with our pooled results because we found no impact of RV test results on rates of antibiotic prescription in bronchiolitis.

In contrast with bronchiolitis guidelines, current recommendations about RV testing for pediatric community-acquired pneumonia in children are vague and inconsistent. The Infectious Diseases Society of America guidelines recommend testing for influenza and suggest testing for other viruses “that can modify clinical decision-making” but do not give additional details.23  The British Thoracic Society (BTS) guidelines recommend testing only in children with complicated community-acquired pneumonia,52  whereas the Canadian Paediatric Society states that testing should be considered in children admitted during influenza season.53  With regards to antibiotics, both the Infectious Diseases Society of America and the BTS recommend empirical antibiotics for hospitalized children; however, the BTS states that children <2 years old should not receive antibiotics because symptoms of respiratory tract infection are not usually pneumonia.23,52  Given these unclear recommendations, it is not surprising that we found high heterogeneity and no overall impact of RV results on antibiotic use.

Although the results of our systematic review and meta-analysis are insufficient to support the routine use of RV testing to reduce antibiotic prescribing, such testing may still be beneficial for other purposes. For example, RV testing can be an important tool to track the prevalence of RVs and to determine the viral etiology of possible nosocomial outbreaks, which can facilitate adequate infection control practices such as patient cohorting. In addition, RV testing can help diagnose influenza and guide antiviral treatment during influenza season in hospitalized patients with severe disease or at a high risk of complications.54 

This study has limitations. We could not obtain sufficient data on the timing of antibiotic prescription and/or ancillary testing with respect to RV test results. We may therefore also be capturing empirical antibiotic use and/or rates of ancillary tests ordered before the release of RV test results, which impairs the ability to study causal relationships between viral testing and these outcomes. However, we believe that authors of prospective cohort studies were able to better observe antibiotic use and/or ancillary testing after RV test results. Furthermore, 20 studies (87%) were at serious risk of bias, most often because of a lack of controlling for disease severity and comorbidities. There was also high statistical heterogeneity among the pooled studies. Despite performing multiple subgroup analyses to control for potential sources of heterogeneity (age, type of RV test, viral target, and a diagnosis of bronchiolitis), statistical heterogeneity remained high within these subgroupings, except when we stratified by study design. Finally, the reporting of age in the included studies was incomplete and inconsistent. To mitigate this problem, we used an operational definition to categorize patient age as infants (≤1 year old) or children (>1 year old). This allowed us to perform an exploratory analysis of age as a potential confounder.

Nevertheless, this is the first systematic review and meta-analysis used to look at the clinical utility of RV tests in children hospitalized with severe ARI, and its results have several implications for future research. Our subgroup analysis revealed a reduction in antibiotic use among prospective cohort studies. Larger such studies are needed to further evaluate the clinical utility of RV testing with control for potential confounders to examine the temporal relationship between viral test results and treatment decisions and evaluate the factors that influence therapeutic decisions.

Importantly, researchers in the studies in our review used immunofluorescence assays, RADTs, or NAAT methods for RV testing. However, in recent years, molecular point-of-care diagnostic tests have been developed that are faster and more accurate than the previous RV testing methods included in our review and have turnaround times ranging from 13 to 60 minutes.55,56  The utility of RV testing may have decreased if results were not available within a time frame that would affect clinicians’ decision-making. Prospective cohort studies are thus needed to evaluate whether the newer, quicker, and more accurate rapid RV tests could impact the management of children with severe ARI.

Lastly, future studies may consider evaluating the impact of RV testing in conjunction with biomarkers and/or antibiotic stewardship programs. Although RV testing may be a reliable method of identifying a virus, it does not rule out a concomitant or secondary bacterial infection. Because diagnostic tests for bacterial infections of the lower respiratory tract are limited and insensitive,57  results from some studies suggest that a combination of viral testing and use of biomarkers can better differentiate viral from bacterial infections and thus more effectively guide antibiotic initiation and/or discontinuation.5861  Moreover, Lowe et al62  showed that a combination of viral testing and antibiotic stewardship recommendations were effective in reducing antibiotic duration in hospitalized adults with viral ARI.

RV testing does not seem to impact the management of children hospitalized with severe ARI in terms of antibiotic use, ancillary testing, or length of stay. Thus, there is currently insufficient evidence to support the routine use of RV testing in this patient population. However, subgroup analysis revealed that prospective cohort studies may better elucidate the impact of RV testing on subsequent antibiotic use and/or ancillary tests. A prospective design and newer RV assays to more thoroughly evaluate the impact of such tests on therapeutic decisions should therefore be used in future studies.

Ms Noël developed the study protocol and search strategy and performed study selection, data collection, data analysis, and manuscript writing; Dr Papenburg developed the study protocol and search strategy, performed critical analysis of the results and critical review of manuscript, and provided expertise on viral diagnostic testing; Dr Fontela developed the study protocol and search strategy and performed critical analysis of the results and critical review of the manuscript; Mr Winters performed study selection, data collection, and critical review of the manuscript; Dr Quach performed critical review of the results and the manuscript and provided expertise on viral diagnostic testing; Dr Dendukuri supervised data analysis and performed critical review of the results and the manuscript; Ms Gore developed the search strategy and performed critical review of the manuscript; Dr Robinson provided expertise on respiratory infections and viral diagnostic testing and performed critical review of the manuscript; and all authors contributed to the development of the selection and data extraction criteria and read, provided feedback on, and approved the final protocol and manuscript as submitted.

FUNDING: Data collection by the original investigators and data management were supported by The Research Institute of the McGill University Health Centre. The Research Institute of the McGill University Health Centre was not involved in any other aspect of this project (eg, protocol design, analysis planning, and statistical analyses). The funder did not have input on the interpretation or publication of the study results.

1
GBD 2015 Mortality and Causes of Death Collaborators
.
Global, regional, and national life expectancy, all-cause mortality, and cause-specific mortality for 249 causes of death, 1980-2015: a systematic analysis for the Global Burden of Disease Study 2015 [published correction appears in Lancet. 2017;389(10064):e1]
.
Lancet
.
2016
;
388
(
10053
):
1459
1544
2
Lafond
KE
,
Nair
H
,
Rasooly
MH
, et al
;
Global Respiratory Hospitalizations—Influenza Proportion Positive (GRIPP) Working Group
.
Global role and burden of influenza in pediatric respiratory hospitalizations, 1982-2012: a systematic analysis [published correction appears in PLoS Med. 2016;13(6):e1002060]
.
PLoS Med
.
2016
;
13
(
3
):
e1001977
3
Jain
S
,
Williams
DJ
,
Arnold
SR
, et al
;
CDC EPIC Study Team
.
Community-acquired pneumonia requiring hospitalization among U.S. children
.
N Engl J Med
.
2015
;
372
(
9
):
835
845
4
Shi
T
,
McAllister
DA
,
O’Brien
KL
, et al
;
RSV Global Epidemiology Network
.
Global, regional, and national disease burden estimates of acute lower respiratory infections due to respiratory syncytial virus in young children in 2015: a systematic review and modelling study
.
Lancet
.
2017
;
390
(
10098
):
946
958
5
Simoes
EAF
,
Cherian
T
,
Chow
J
,
Shahid-Salles
SA
,
Laxminarayan
R
,
John
TJ
.
Acute respiratory infections in children
. In:
Jamison
DT
,
Breman
JG
,
Measham
AR
, et al
, eds.
Disease Control Priorities in Developing Countries
. 2nd ed.
Washington, DC; New York, NY
:
The International Bank for Reconstruction and Development/The World Bank/Oxford University Press
;
2006
:
483
498
6
Bellos
A
,
Mulholland
K
,
O’Brien
KL
,
Qazi
SA
,
Gayer
M
,
Checchi
F
.
The burden of acute respiratory infections in crisis-affected populations: a systematic review
.
Confl Health
.
2010
;
4
:
3
7
Mullins
JA
,
Lamonte
AC
,
Bresee
JS
,
Anderson
LJ
.
Substantial variability in community respiratory syncytial virus season timing
.
Pediatr Infect Dis J
.
2003
;
22
(
10
):
857
862
8
Langley
JM
,
Wang
EE
,
Law
BJ
, et al
.
Economic evaluation of respiratory syncytial virus infection in Canadian children: a Pediatric Investigators Collaborative Network on Infections in Canada (PICNIC) study
.
J Pediatr
.
1997
;
131
(
1, pt 1
):
113
117
9
Hall
CB
,
Weinberg
GA
,
Iwane
MK
, et al
.
The burden of respiratory syncytial virus infection in young children
.
N Engl J Med
.
2009
;
360
(
6
):
588
598
10
Farley
R
,
Spurling
GK
,
Eriksson
L
,
Del Mar
CB
.
Antibiotics for bronchiolitis in children under two years of age
.
Cochrane Database Syst Rev
.
2014
;(
10
):
CD005189
11
Samson
L
,
Cooke
C
,
Macdonald
N
.
Analysis of antibiotic use and misuse in children hospitalized with RSV infection
.
Paediatr Child Health
.
1999
;
4
(
3
):
195
199
12
Henderson
M
,
Rubin
E
.
Misuse of antimicrobials in children with asthma and bronchiolitis: a review
.
Pediatr Infect Dis J
.
2001
;
20
(
2
):
214
215
13
Rogers
BB
,
Shankar
P
,
Jerris
RC
, et al
.
Impact of a rapid respiratory panel test on patient outcomes
.
Arch Pathol Lab Med
.
2015
;
139
(
5
):
636
641
14
Archimbaud
C
,
Ouchchane
L
,
Mirand
A
, et al
.
Improvement of the management of infants, children and adults with a molecular diagnosis of Enterovirus meningitis during two observational study periods
.
PLoS One
.
2013
;
8
(
7
):
e68571
15
Gillies
M
,
Ranakusuma
A
,
Hoffmann
T
, et al
.
Common harms from amoxicillin: a systematic review and meta-analysis of randomized placebo-controlled trials for any indication
.
CMAJ
.
2015
;
187
(
1
):
E21
E31
16
Shehab
N
,
Patel
PR
,
Srinivasan
A
,
Budnitz
DS
.
Emergency department visits for antibiotic-associated adverse events
.
Clin Infect Dis
.
2008
;
47
(
6
):
735
743
17
Bradshaw
ML
,
Déragon
A
,
Puligandla
P
,
Émeriaud
G
,
Canakis
AM
,
Fontela
PS
.
Treatment of severe bronchiolitis: a survey of Canadian pediatric intensivists
.
Pediatr Pulmonol
.
2018
;
53
(
5
):
613
618
18
Gill
PJ
,
Richardson
SE
,
Ostrow
O
,
Friedman
JN
.
Testing for respiratory viruses in children: to swab or not to swab
.
JAMA Pediatr
.
2017
;
171
(
8
):
798
804
19
Woo
PC
,
Chiu
SS
,
Seto
WH
,
Peiris
M
.
Cost-effectiveness of rapid diagnosis of viral respiratory tract infections in pediatric patients
.
J Clin Microbiol
.
1997
;
35
(
6
):
1579
1581
20
Mills
JM
,
Harper
J
,
Broomfield
D
,
Templeton
KE
.
Rapid testing for respiratory syncytial virus in a paediatric emergency department: benefits for infection control and bed management
.
J Hosp Infect
.
2011
;
77
(
3
):
248
251
21
Thibeault
R
,
Gilca
R
,
Côté
S
,
De Serres
G
,
Boivin
G
,
Déry
P
.
Antibiotic use in children is not influenced by the result of rapid antigen detection test for the respiratory syncytial virus
.
J Clin Virol
.
2007
;
39
(
3
):
169
174
22
Barenfanger
J
,
Drake
C
,
Leon
N
,
Mueller
T
,
Troutt
T
.
Clinical and financial benefits of rapid detection of respiratory viruses: an outcomes study
.
J Clin Microbiol
.
2000
;
38
(
8
):
2824
2828
23
Bradley
JS
,
Byington
CL
,
Shah
SS
, et al
;
Pediatric Infectious Diseases Society and the Infectious Diseases Society of America
.
The management of community-acquired pneumonia in infants and children older than 3 months of age: clinical practice guidelines by the Pediatric Infectious Diseases Society and the Infectious Diseases Society of America
.
Clin Infect Dis
.
2011
;
53
(
7
):
e25
e76
24
Le Saux
N
,
Robinson
JL
;
Canadian Paediatric Society, Infectious Diseases and Immunization Committee
.
Uncomplicated pneumonia in healthy Canadian children and youth: practice points for management
.
Paediatr Child Health
.
2015
;
20
(
8
):
441
450
25
Sterne
JA
,
Hernán
MA
,
Reeves
BC
, et al
.
ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions
.
BMJ
.
2016
;
355
:
i4919
26
Wells
GA
,
Shea
B
,
O’Connell
D
, et al
;
The Ottawa Hospital Research Institute
.
The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses
.
2013
.
Available at: www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed June 1, 2018
27
Chartrand
C
,
Tremblay
N
,
Renaud
C
,
Papenburg
J
.
Diagnostic accuracy of rapid antigen detection tests for respiratory syncytial virus infection: systematic review and meta-analysis
.
J Clin Microbiol
.
2015
;
53
(
12
):
3738
3749
28
R Core Team
.
R: A Language and Environment for Statistical Computing
.
Vienna, Austria
:
R Foundation for Statistical Computing
;
2013
29
Adcock
PM
,
Stout
GG
,
Hauck
MA
,
Marshall
GS
.
Effect of rapid viral diagnosis on the management of children hospitalized with lower respiratory tract infection
.
Pediatr Infect Dis J
.
1997
;
16
(
9
):
842
846
30
Azzarone
G
,
Agalliu
I
,
Nazif
J
,
Esteban
N
,
Biller
R
,
Liewehr
S
.
RSV antigen testing: clinically relevant or mere academic exercise?
In:
Hospital Medicine
;
April 1–4, 2012
;
San Diego, CA
31
Bozdemir
SE
,
Çelebi
S
,
Çakir
D
, et al
.
Direct medical cost assessment in the <2 years old hospitalized RSV+LRTI patients
.
J Pediatr Inf
.
2016
;
10
(
4
):
128
136
32
Bueno
IA
,
Riccetto
AG
,
Morcillo
AM
,
Arns
CW
,
Baracat
EC
.
Respiratory syncytial virus, infants and intensive therapy
.
Braz J Infect Dis
.
2012
;
16
(
1
):
86
89
33
Byington
CL
,
Castillo
H
,
Gerber
K
, et al
.
The effect of rapid respiratory viral diagnostic testing on antibiotic use in a children’s hospital
.
Arch Pediatr Adolesc Med
.
2002
;
156
(
12
):
1230
1234
34
Ferronato
ÂE
,
Gilio
AE
,
Ferraro
AA
,
Paulis
Md
,
Vieira
SE
.
Etiological diagnosis reduces the use of antibiotics in infants with bronchiolitis
.
Clinics (São Paulo)
.
2012
;
67
(
9
):
1001
1006
35
Flaherman
V
,
Li
S
,
Ragins
A
,
Masaquel
A
,
Kipnis
P
,
Escobar
GJ
.
Respiratory syncytial virus testing during bronchiolitis episodes of care in an integrated health care delivery system: a retrospective cohort study
.
Clin Ther
.
2010
;
32
(
13
):
2220
2229
36
Hatipoğlu
N
,
Somer
A
,
Badur
S
, et al
.
Viral etiology in hospitalized children with acute lower respiratory tract infection
.
Turk J Pediatr
.
2011
;
53
(
5
):
508
516
37
Manji
R
,
Lotlikar
M
,
Zhang
F
,
Ginocchio
CC
.
Clinical evaluation of NucliSENS magnetic extraction and NucliSENS analytical specific reagents for the real-time detection of respiratory syncytial virus (RSV) in paediatric respiratory specimens
.
J Clin Pathol
.
2009
;
62
(
11
):
998
1002
38
McCulloh
RJ
,
Andrea
S
,
Reinert
S
,
Chapin
K
.
Potential utility of multiplex amplification respiratory viral panel testing in the management of acute respiratory infection in children: a retrospective analysis
.
J Pediatric Infect Dis Soc
.
2014
;
3
(
2
):
146
153
39
Milić
P
,
Sikirica
M
,
Krželj
V
,
Markić
J
.
Characteristics of infants hospitalized with bronchiolitis at University Hospital of Split between 2011 and 2015
.
Paediatr Croat
.
2017
;
61
(
2
):
53
58
40
Nitsch-Osuch
A
,
Gyrczuk
E
,
Wardyn
A
,
Życinska
K
,
Brydak
L
.
Antibiotic prescription practices among children with influenza
.
Adv Exp Med Biol
.
2016
;
905
:
25
31
41
Paul
SP
,
Mukherjee
A
,
McAllister
T
,
Harvey
MJ
,
Clayton
BA
,
Turner
PC
.
Respiratory-syncytial-virus- and rhinovirus-related bronchiolitis in children aged <2 years in an English district general hospital
.
J Hosp Infect
.
2017
;
96
(
4
):
360
365
42
Schulert
GS
,
Lu
Z
,
Wingo
T
,
Tang
YW
,
Saville
BR
,
Hain
PD
.
Role of a respiratory viral panel in the clinical management of pediatric inpatients
.
Pediatr Infect Dis J
.
2013
;
32
(
5
):
467
472
43
Schulert
GS
,
Hain
PD
,
Williams
DJ
.
Utilization of viral molecular diagnostics among children hospitalized with community acquired pneumonia
.
Hosp Pediatr
.
2014
;
4
(
6
):
372
376
44
Sulieman
SE
,
Herigon
JC
,
Newland
JG
,
Selvarangan
R
.
Impact of rapid viral cultures on management of hospitalized children with respiratory infections
. In:
Infectious Diseases Society of America Annual Conference
;
October 29–November 1, 2009
;
Philadelphia, PA
45
Suntarattiwong
P
,
Sojisirikul
K
,
Sitaposa
P
, et al
.
Clinical and epidemiological characteristics of respiratory syncytial virus and influenza virus associated hospitalization in urban Thai infants
.
J Med Assoc Thai
.
2011
;
94
(
suppl 3
):
S164
S171
46
Tresoldi
AT
,
Pereira
RM
,
Fraga
AMA
, et al
.
Clinical features and outcome of children and adolescents hospitalized with influenza A (H1N1) virus infection compared with flu-like symptoms and negative rapid tests for influenza A (H1N1) admitted in the same period of time
.
J Trop Pediatr
.
2011
;
57
(
6
):
481
483
47
Tsolia
MN
,
Kafetzis
D
,
Danelatou
K
, et al
.
Epidemiology of respiratory syncytial virus bronchiolitis in hospitalized infants in Greece
.
Eur J Epidemiol
.
2003
;
18
(
1
):
55
61
48
Tsung
LY
,
Choi
KC
,
Nelson
EA
,
Chan
PK
,
Sung
RY
.
Factors associated with length of hospital stay in children with respiratory disease
.
Hong Kong Med J
.
2010
;
16
(
6
):
440
446
49
van de Pol
AC
,
Wolfs
TF
,
Tacke
CE
, et al
.
Impact of PCR for respiratory viruses on antibiotic use: theory and practice
.
Pediatr Pulmonol
.
2011
;
46
(
5
):
428
434
50
Walls
T
,
Stark
E
,
Pattemore
P
,
Jennings
L
.
Missed opportunities for antimicrobial stewardship in pre-school children admitted to hospital with lower respiratory tract infection
.
J Paediatr Child Health
.
2017
;
53
(
6
):
569
571
51
American Academy of Pediatrics Subcommittee on Diagnosis and Management of Bronchiolitis
.
Diagnosis and management of bronchiolitis
.
Pediatrics
.
2006
;
118
(
4
):
1774
1793
52
Harris
M
,
Clark
J
,
Coote
N
, et al
;
British Thoracic Society Standards of Care Committee
.
British Thoracic Society guidelines for the management of community acquired pneumonia in children: update 2011
.
Thorax
.
2011
;
66
(
suppl 2
):
ii1
ii23
53
Le Saux
N
,
Robinson
J
.
Pneumonia in healthy Canadian children and youth: practice points for management
.
Paediatr Child Health
.
2011
;
16
(
7
):
417
424
54
Uyeki
TM
,
Bernstein
HH
,
Bradley
JS
, et al
.
Clinical practice guidelines by the Infectious Diseases Society of America: 2018 update on diagnosis, treatment, chemoprophylaxis, and institutional outbreak management of seasonal influenza
.
Clin Infect Dis
.
2019
;
68
(
6
):
e1
e47
55
Hogan
CA
,
Caya
C
,
Papenburg
J
.
Rapid and simple molecular tests for the detection of respiratory syncytial virus: a review
.
Expert Rev Mol Diagn
.
2018
;
18
(
7
):
617
629
56
Merckx
J
,
Wali
R
,
Schiller
I
, et al
.
Diagnostic accuracy of novel and traditional rapid tests for influenza infection compared with reverse transcriptase polymerase chain reaction: a systematic review and meta-analysis
.
Ann Intern Med
.
2017
;
167
(
6
):
394
409
57
Carroll
KC
.
Laboratory diagnosis of lower respiratory tract infections: controversy and conundrums
.
J Clin Microbiol
.
2002
;
40
(
9
):
3115
3120
58
Yusa
T
,
Tateda
K
,
Ohara
A
,
Miyazaki
S
.
New possible biomarkers for diagnosis of infections and diagnostic distinction between bacterial and viral infections in children
.
J Infect Chemother
.
2017
;
23
(
2
):
96
100
59
Branche
AR
,
Walsh
EE
,
Vargas
R
, et al
.
Serum procalcitonin measurement and viral testing to guide antibiotic use for respiratory infections in hospitalized adults: a randomized controlled trial
.
J Infect Dis
.
2015
;
212
(
11
):
1692
1700
60
Joseph
P
,
Godofsky
E
.
Outpatient antibiotic stewardship: a growing frontier-combining myxovirus resistance protein a with other biomarkers to improve antibiotic use
.
Open Forum Infect Dis
.
2018
;
5
(
2
):
ofy024
61
Fontela
PS
,
O’Donnell
S
,
Papenburg
J
.
Can biomarkers improve the rational use of antibiotics?
Curr Opin Infect Dis
.
2018
;
31
(
4
):
347
352
62
Lowe
CF
,
Payne
M
,
Puddicombe
D
, et al
.
Antimicrobial stewardship for hospitalized patients with viral respiratory tract infections
.
Am J Infect Control
.
2017
;
45
(
8
):
872
875

Competing Interests

POTENTIAL CONFLICT OF INTEREST: Dr Papenburg acknowledges receiving consulting and/or honoraria fees or research grant funding outside of the current work from the following: AbbVie, BD Diagnostics, Cepheid, and MedImmune; the other authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data