Advertising Disclaimer »

Review Article

The Clinical Utility of Respiratory Viral Testing in Hospitalized Children: A Meta-analysis

Kim C. Noël, Patricia S. Fontela, Nicholas Winters, Caroline Quach, Genevieve Gore, Joan Robinson, Nandini Dendukuri and Jesse Papenburg
Hospital Pediatrics July 2019, 9 (7) 483-494; DOI: https://doi.org/10.1542/hpeds.2018-0233
Download PDF

Abstract

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.1–3 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.7–9

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.10–12 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.13,18–22 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.

Methods

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).

Information Sources and Search Strategy

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.

Eligibility Criteria

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).

Study Selection

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

Data Collection Process

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.

Quality Assessment

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

Statistical Analysis

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

Results

Study Selection

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).21,29–50

View this table:
TABLE 1

Characteristics of Included Studies

Study Characteristics

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.

Quality of Included 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).

Antibiotic Prescribing

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).29,35–38,40,41 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%).30–33,39,42,44,45,47–49 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
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
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
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
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).

Duration of Antibiotic Use

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.

Antibiotic Discontinuation

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).

Ordering of Ancillary Tests

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).

Hospital Length of Stay

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.

Use of Antiviral Agents

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).

Risk of Publication Bias

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

Discussion

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.58–61 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.

Conclusions

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.

Footnotes

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

  • 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.

  • 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.

References

  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
    OpenUrlCrossRefPubMed
    1. Lafond KE,
    2. Nair H,
    3. Rasooly MH,
    4. 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
    OpenUrlCrossRefPubMed
  3. ↵
    1. Jain S,
    2. Williams DJ,
    3. Arnold SR,
    4. et al
    ; CDC EPIC Study Team. Community-acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med. 2015;372(9):835–845
    OpenUrlCrossRefPubMed
  4. ↵
    1. Shi T,
    2. McAllister DA,
    3. O’Brien KL,
    4. 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
    OpenUrlCrossRefPubMed
  5. ↵
    1. Simoes EAF,
    2. Cherian T,
    3. Chow J,
    4. Shahid-Salles SA,
    5. Laxminarayan R,
    6. 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. ↵
    1. Bellos A,
    2. Mulholland K,
    3. O’Brien KL,
    4. Qazi SA,
    5. Gayer M,
    6. Checchi F
    . The burden of acute respiratory infections in crisis-affected populations: a systematic review. Confl Health. 2010;4:3
    OpenUrlCrossRefPubMed
  7. ↵
    1. Mullins JA,
    2. Lamonte AC,
    3. Bresee JS,
    4. Anderson LJ
    . Substantial variability in community respiratory syncytial virus season timing. Pediatr Infect Dis J. 2003;22(10):857–862
    OpenUrlCrossRefPubMed
    1. Langley JM,
    2. Wang EE,
    3. Law BJ,
    4. 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
    OpenUrlCrossRefPubMed
  9. ↵
    1. Hall CB,
    2. Weinberg GA,
    3. Iwane MK,
    4. et al
    . The burden of respiratory syncytial virus infection in young children. N Engl J Med. 2009;360(6):588–598
    OpenUrlCrossRefPubMed
  10. ↵
    1. Farley R,
    2. Spurling GK,
    3. Eriksson L,
    4. Del Mar CB
    . Antibiotics for bronchiolitis in children under two years of age. Cochrane Database Syst Rev. 2014;(10):CD005189
    1. Samson L,
    2. Cooke C,
    3. Macdonald N
    . Analysis of antibiotic use and misuse in children hospitalized with RSV infection. Paediatr Child Health. 1999;4(3):195–199
    OpenUrlPubMed
  12. ↵
    1. Henderson M,
    2. Rubin E
    . Misuse of antimicrobials in children with asthma and bronchiolitis: a review. Pediatr Infect Dis J. 2001;20(2):214–215
    OpenUrlPubMed
  13. ↵
    1. Rogers BB,
    2. Shankar P,
    3. Jerris RC,
    4. et al
    . Impact of a rapid respiratory panel test on patient outcomes. Arch Pathol Lab Med. 2015;139(5):636–641
    OpenUrlCrossRefPubMed
  14. ↵
    1. Archimbaud C,
    2. Ouchchane L,
    3. Mirand A,
    4. 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
    OpenUrlCrossRefPubMed
  15. ↵
    1. Gillies M,
    2. Ranakusuma A,
    3. Hoffmann T,
    4. 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
    OpenUrlAbstract/FREE Full Text
  16. ↵
    1. Shehab N,
    2. Patel PR,
    3. Srinivasan A,
    4. Budnitz DS
    . Emergency department visits for antibiotic-associated adverse events. Clin Infect Dis. 2008;47(6):735–743
    OpenUrlCrossRefPubMed
  17. ↵
    1. Bradshaw ML,
    2. Déragon A,
    3. Puligandla P,
    4. Émeriaud G,
    5. Canakis AM,
    6. Fontela PS
    . Treatment of severe bronchiolitis: a survey of Canadian pediatric intensivists. Pediatr Pulmonol. 2018;53(5):613–618
    OpenUrl
  18. ↵
    1. Gill PJ,
    2. Richardson SE,
    3. Ostrow O,
    4. Friedman JN
    . Testing for respiratory viruses in children: to swab or not to swab. JAMA Pediatr. 2017;171(8):798–804
    OpenUrl
    1. Woo PC,
    2. Chiu SS,
    3. Seto WH,
    4. Peiris M
    . Cost-effectiveness of rapid diagnosis of viral respiratory tract infections in pediatric patients. J Clin Microbiol. 1997;35(6):1579–1581
    OpenUrlAbstract/FREE Full Text
    1. Mills JM,
    2. Harper J,
    3. Broomfield D,
    4. 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
    OpenUrlCrossRefPubMed
  21. ↵
    1. Thibeault R,
    2. Gilca R,
    3. Côté S,
    4. De Serres G,
    5. Boivin G,
    6. 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
    OpenUrlCrossRefPubMed
  22. ↵
    1. Barenfanger J,
    2. Drake C,
    3. Leon N,
    4. Mueller T,
    5. Troutt T
    . Clinical and financial benefits of rapid detection of respiratory viruses: an outcomes study. J Clin Microbiol. 2000;38(8):2824–2828
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Bradley JS,
    2. Byington CL,
    3. Shah SS,
    4. 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
    OpenUrlCrossRefPubMed
  24. ↵
    1. Le Saux N,
    2. 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
    OpenUrl
  25. ↵
    1. Sterne JA,
    2. Hernán MA,
    3. Reeves BC,
    4. et al
    . ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ. 2016;355:i4919
    OpenUrlFREE Full Text
  26. ↵
    1. Wells GA,
    2. Shea B,
    3. O’Connell D,
    4. 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. ↵
    1. Chartrand C,
    2. Tremblay N,
    3. Renaud C,
    4. 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
    OpenUrlAbstract/FREE Full Text
  28. ↵
    R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing; 2013
  29. ↵
    1. Adcock PM,
    2. Stout GG,
    3. Hauck MA,
    4. 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
    OpenUrlCrossRefPubMed
  30. ↵
    1. Azzarone G,
    2. Agalliu I,
    3. Nazif J,
    4. Esteban N,
    5. Biller R,
    6. Liewehr S
    . RSV antigen testing: clinically relevant or mere academic exercise? In: Hospital Medicine; April 1–4, 2012; San Diego, CA
  31. ↵
    1. Bozdemir SE,
    2. Çelebi S,
    3. Çakir D,
    4. et al
    . Direct medical cost assessment in the <2 years old hospitalized RSV+LRTI patients. J Pediatr Inf. 2016;10(4):128–136
    OpenUrl
    1. Bueno IA,
    2. Riccetto AG,
    3. Morcillo AM,
    4. Arns CW,
    5. Baracat EC
    . Respiratory syncytial virus, infants and intensive therapy. Braz J Infect Dis. 2012;16(1):86–89
    OpenUrlCrossRefPubMed
  33. ↵
    1. Byington CL,
    2. Castillo H,
    3. Gerber K,
    4. 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
    OpenUrlCrossRefPubMed
  34. ↵
    1. Ferronato ÂE,
    2. Gilio AE,
    3. Ferraro AA,
    4. Paulis Md,
    5. Vieira SE
    . Etiological diagnosis reduces the use of antibiotics in infants with bronchiolitis. Clinics (São Paulo). 2012;67(9):1001–1006
    OpenUrlCrossRefPubMed
  35. ↵
    1. Flaherman V,
    2. Li S,
    3. Ragins A,
    4. Masaquel A,
    5. Kipnis P,
    6. 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
    OpenUrlCrossRefPubMed
    1. Hatipoğlu N,
    2. Somer A,
    3. Badur S,
    4. et al
    . Viral etiology in hospitalized children with acute lower respiratory tract infection. Turk J Pediatr. 2011;53(5):508–516
    OpenUrlPubMed
  37. ↵
    1. Manji R,
    2. Lotlikar M,
    3. Zhang F,
    4. 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
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. McCulloh RJ,
    2. Andrea S,
    3. Reinert S,
    4. 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
    OpenUrlCrossRefPubMed
  39. ↵
    1. Milić P,
    2. Sikirica M,
    3. Krželj V,
    4. Markić J
    . Characteristics of infants hospitalized with bronchiolitis at University Hospital of Split between 2011 and 2015. Paediatr Croat. 2017;61(2):53–58
    OpenUrl
  40. ↵
    1. Nitsch-Osuch A,
    2. Gyrczuk E,
    3. Wardyn A,
    4. Życinska K,
    5. Brydak L
    . Antibiotic prescription practices among children with influenza. Adv Exp Med Biol. 2016;905:25–31
    OpenUrl
  41. ↵
    1. Paul SP,
    2. Mukherjee A,
    3. McAllister T,
    4. Harvey MJ,
    5. Clayton BA,
    6. 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
    OpenUrl
  42. ↵
    1. Schulert GS,
    2. Lu Z,
    3. Wingo T,
    4. Tang YW,
    5. Saville BR,
    6. Hain PD
    . Role of a respiratory viral panel in the clinical management of pediatric inpatients. Pediatr Infect Dis J. 2013;32(5):467–472
    OpenUrlCrossRefPubMed
  43. ↵
    1. Schulert GS,
    2. Hain PD,
    3. Williams DJ
    . Utilization of viral molecular diagnostics among children hospitalized with community acquired pneumonia. Hosp Pediatr. 2014;4(6):372–376
    OpenUrlAbstract/FREE Full Text
  44. ↵
    1. Sulieman SE,
    2. Herigon JC,
    3. Newland JG,
    4. 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. ↵
    1. Suntarattiwong P,
    2. Sojisirikul K,
    3. Sitaposa P,
    4. 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
    OpenUrl
  46. ↵
    1. Tresoldi AT,
    2. Pereira RM,
    3. Fraga AMA,
    4. 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
    OpenUrlCrossRefPubMed
  47. ↵
    1. Tsolia MN,
    2. Kafetzis D,
    3. Danelatou K,
    4. et al
    . Epidemiology of respiratory syncytial virus bronchiolitis in hospitalized infants in Greece. Eur J Epidemiol. 2003;18(1):55–61
    OpenUrlPubMed
  48. ↵
    1. Tsung LY,
    2. Choi KC,
    3. Nelson EA,
    4. Chan PK,
    5. Sung RY
    . Factors associated with length of hospital stay in children with respiratory disease. Hong Kong Med J. 2010;16(6):440–446
    OpenUrlPubMed
  49. ↵
    1. van de Pol AC,
    2. Wolfs TF,
    3. Tacke CE,
    4. et al
    . Impact of PCR for respiratory viruses on antibiotic use: theory and practice. Pediatr Pulmonol. 2011;46(5):428–434
    OpenUrl
  50. ↵
    1. Walls T,
    2. Stark E,
    3. Pattemore P,
    4. 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
    OpenUrl
  51. ↵
    American Academy of Pediatrics Subcommittee on Diagnosis and Management of Bronchiolitis. Diagnosis and management of bronchiolitis. Pediatrics. 2006;118(4):1774–1793
    OpenUrlAbstract/FREE Full Text
  52. ↵
    1. Harris M,
    2. Clark J,
    3. Coote N,
    4. 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
    OpenUrlAbstract/FREE Full Text
  53. ↵
    1. Le Saux N,
    2. Robinson J
    . Pneumonia in healthy Canadian children and youth: practice points for management. Paediatr Child Health. 2011;16(7):417–424
    OpenUrlPubMed
  54. ↵
    1. Uyeki TM,
    2. Bernstein HH,
    3. Bradley JS,
    4. 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
    OpenUrl
  55. ↵
    1. Hogan CA,
    2. Caya C,
    3. 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
    OpenUrl
  56. ↵
    1. Merckx J,
    2. Wali R,
    3. Schiller I,
    4. 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
    OpenUrl
  57. ↵
    1. Carroll KC
    . Laboratory diagnosis of lower respiratory tract infections: controversy and conundrums. J Clin Microbiol. 2002;40(9):3115–3120
    OpenUrlFREE Full Text
  58. ↵
    1. Yusa T,
    2. Tateda K,
    3. Ohara A,
    4. 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
    OpenUrl
    1. Branche AR,
    2. Walsh EE,
    3. Vargas R,
    4. 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
    OpenUrlCrossRefPubMed
    1. Joseph P,
    2. 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
    OpenUrl
  61. ↵
    1. Fontela PS,
    2. O’Donnell S,
    3. Papenburg J
    . Can biomarkers improve the rational use of antibiotics? Curr Opin Infect Dis. 2018;31(4):347–352
    OpenUrl
  62. ↵
    1. Lowe CF,
    2. Payne M,
    3. Puddicombe D,
    4. et al
    . Antimicrobial stewardship for hospitalized patients with viral respiratory tract infections. Am J Infect Control. 2017;45(8):872–875
    OpenUrlCrossRef
View Abstract
Back to top

In this issue

View this article with LENS
Email Article
Request Permissions
Article Alerts
Citation Tools
The Clinical Utility of Respiratory Viral Testing in Hospitalized Children: A Meta-analysis
Kim C. Noël, Patricia S. Fontela, Nicholas Winters, Caroline Quach, Genevieve Gore, Joan Robinson, Nandini Dendukuri, Jesse Papenburg
Hospital Pediatrics Jul 2019, 9 (7) 483-494; DOI: 10.1542/hpeds.2018-0233
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
Print
Download PDF
Insight Alerts

Jump to section

Subjects