Secondary Logo

Journal Logo

Original Studies

Antibiotic Overuse in Children with Respiratory Syncytial Virus Lower Respiratory Tract Infection

van Houten, Chantal B. MD*; Naaktgeboren, Christiana PhD; Buiteman, Brigitte J. M. MSc*; van der Lee, Maaike MSc*; Klein, Adi MD; Srugo, Isaac MD§; Chistyakov, Irena MD§; de Waal, Wouter PhD; Meijssen, Clemens B. MD; Meijers, Pieter W. MD**; de Winter-de Groot, Karin M. MD††; Wolfs, Tom F.W. PhD*; Shachor-Meyouhas, Yael MD‡‡; Stein, Michal MD§§; Sanders, Elisabeth A. M. PhD*; Bont, Louis J. PhD*

Author Information
The Pediatric Infectious Disease Journal: November 2018 - Volume 37 - Issue 11 - p 1077-1081
doi: 10.1097/INF.0000000000001981

Abstract

Respiratory syncytial virus (RSV) is the most common cause of viral lower respiratory tract infections (LRTI) during the first year of life. In winter seasons, 20% of hospitalizations and 18% of emergency department visits for acute respiratory infections of young children is associated with RSV infection.1 A Cochrane review found insufficient evidence to support the use of antibiotics for bronchiolitis.2 Accordingly, antibiotics are not recommended for RSV LRTI unless there is clear evidence of secondary bacterial disease or respiratory failure.2

There is a large variation in the reported percentages of bacterial coinfection in RSV patients, though most studies report less than 11%.3–10 Despite the low incidence of bacterial coinfections in patients infected with RSV, antibiotic prescriptions range between 29% and 80%.3,4,9–12 Antibiotic overuse is associated with an expansion in antibiotic resistance. Yearly the numbers of patients dying as a direct result of infections caused by antibiotic-resistant micro-organisms in the United States and Europe are 23,000 and 25,000, respectively. The total economic burden is estimated to be $20 billion direct health care costs and $35 billion a year additional costs for lost productivity in the United States.13,14

The aim of the current prospective, international multicenter study, was to investigate the frequency of bacterial coinfections, and antibiotic overuse in children with RSV infections using expert panel diagnosis on bacterial causes of disease as the reference standard.

METHODS

Setting and Participants

The patient population consisted of 2 similar prospective cohort studies, OPPORTUNITY and TAILORED. The aim of both studies was to differentiate bacterial from viral infections. Participants of OPPORTUNITY were described previously.15 Patient recruitment for both studies took place at the emergency department and ward of 3 secondary and 1 university hospital in the Netherlands and 2 secondary hospitals in Israel. In both studies, children 1 month of age and older presenting with fever (peak temperature ≥ 38.0°C) or symptoms of LRTI were eligible for inclusion. Patients were excluded in case of a previous episode of fever in the past 3 weeks, psychomotor retardation, moderate-to-severe metabolic disorder, primary or secondary immunodeficiency, HIV-, HBV- or HCV infection and active malignancies. Antibiotic use before inclusion was noted. Parents gave written informed consent before sampling. The studies are registered on ClinicalTrials.gov, NCT01931254 and NCT02025699 and were approved by the ethics committees in the participating countries.

Data Collection

Within 24 hours after presentation, a nasal swab (Universal Transport Medium, Copan) was collected for multiplex RT-PCR testing of the most common respiratory pathogens, including RSV and stored at -80°C. The swabs were performed for research and were analyzed when patient recruitment was finished; therefore, the attending physician was not informed about the results of the nasal swab. Samples from the OPPORTUNITY study were transferred to MeMed Diagnostics (Tirat Carmel, Israel) (RV15 test kit, Seegene, KR), samples from the TAILORED study were transferred to the laboratory of Utrecht University Hospital (Magna Pure 96 DNA and Viral NA Large Volume and Magna Pure LC Total Nucleic Acid Isolation Kit, both Roche). After 28 days, we took a questionnaire for clinical follow-up. Clinical data were collected from medical records using an electronic Case Report Form (eCRF). The eCRF consisted of demographics, medical history, physical examination findings, radiologic and laboratory data, microbiologic test results (including study-specific nasal swab data) and follow-up information.

Reference Outcome

Currently, no single reference standard test exists to determine the etiology of an infection.16 Therefore, we followed the England’s National Health Service’s standard for evaluating diagnostic tests and composed an expert panel reference standard.17 We have recently reported on the use of an expert panel as reference standard.15 The panel comprised pediatricians with at least 10 years of clinical experience. Every recruited patient was diagnosed by 3 panel members using all available eCRF information, including the 28-day follow-up, but blinded for the labels of their peers. Each expert assigned 1 of the following etiologies to each patient: bacterial infection, viral infection, bacterial coinfection (i.e., viral and bacterial coinfection) or indeterminate. When at least 2 panel members assigned the same etiology, that etiology was considered to be present. Patients were assigned an inconclusive outcome if each panel member assigned a different etiology or when at least 2 panel members diagnosed the patients as indeterminate. By design, these patients could not be used for the objectives of this study.

Statistical Analysis

Children were first stratified according to the result of the Polymerase Chain Reaction (PCR) test for RSV. Thereafter, we stratified the children according to the reference diagnosis (i.e., simple viral, bacterial coinfection or inconclusive) and studied the use of antibiotic treatment for every reference standard outcome separately. Subcohort analyses were performed on patients admitted to the pediatric intensive care unit (PICU), per age group and for the Dutch and Israeli cohort separately. For baseline characteristics, univariate comparisons were performing Fisher exact test, t tests and Mann-Whitney test as appropriate. Statistical analysis was performed by using SPSS version 21.0 for Windows (IBM Corp., Armonk, NY). A P value < 0.05 was considered statistically significant.

RESULTS

Patient Characteristics

A total of 784 children were recruited between October 2013 and May 2016 (Fig. 1). In 188 (24%) patients, the nasal swab was positive for RSV. The median age of RSV patients was significantly lower when compared with RSV-negative patients, table S1, https://links.lww.com/INF/D26. Of these RSV patients, median age was 9 months, and the majority (60%) of the patients were boys (Table 1). As shown in table S2a, https://links.lww.com/INF/D27, there were some differences between RSV patients recruited in the different countries. When comparing patients recruited in secondary care centers only, the cohorts were more similar (table S2b, https://links.lww.com/INF/D27). In RSV patients with and without antibiotic treatment, many other viruses were detected, rhinovirus was most common (table S3, https://links.lww.com/INF/D28).

TABLE 1.
TABLE 1.:
Baseline Characteristics RSV Cohort
FIGURE 1.
FIGURE 1.:
Recruitment and flow of children presenting at the hospital with fever without source and acute respiratory tract infections. *Inconclusive expert panel diagnosis was assigned to patients where each panel member gave a different diagnosis or where 2 or more members diagnosed the patient as indeterminate. AB, antibiotics.

Bacterial Co-infection during RSV LRTI

The expert panel assigned reference standard outcomes to all 188 RSV patients (Fig. 1) of which 27 (14%) were diagnosed with bacterial coinfection. In 10 (5%) patients, the reference standard outcome was inconclusive; clinical characteristics are shown in Table 1. At presentation, patients with bacterial coinfection appeared more ill than RSV patients without bacterial coinfection (P < 0.001; Table 1). RSV patients with bacterial coinfection were significantly more often admitted at the PICU compared with patients diagnosed with viral infection (63% vs. 7%, respectively; P < 0.001). When looking at the cohort in which all 3 expert panels agree on the diagnosis, (unanimous cohort) 16 (9%) of the patients was diagnosed with a bacterial coinfection (table S5, https://links.lww.com/INF/D251).

Use of Antibiotics

Of all 188 RSV patients, 92 (49%) patients were treated with antibiotics (Fig. 1). Supplementary table 4, https://links.lww.com/INF/D29 shows that amoxicillin/clavulanate was prescribed most often in the Netherlands; in Israel, most children received amoxicillin. All 27 RSV patients with a bacterial coinfection were treated with antibiotics. The most common bacterial diagnoses are shown in Table 1. A bacterial pathogen was detected in only 5 (19%) of the patients. Antibiotics were also administered to 57 (30%) RSV patients with viral infection as reference standard outcome (Fig. 1). Of them, 54 (95%) received antimicrobial treatment within 72 hours after presentation at the hospital (Fig. 1). Even when using our more strict unanimous reference standard, 40 patients (21%) without bacterial coinfection received antibiotics (table S3, https://links.lww.com/INF/D28). Off all RSV patients with a simple viral LRTI receiving antibiotics, 15 (33%) patients had pneumonia and 29 (63%) bronchiolitis as clinical syndrome. Eight RSV children with viral upper respiratory tract infection received also antibiotics (1 case of tonsillitis, 7 cases with nonspecified upper respiratory tract infection). In RSV patients without a bacterial coinfection, patients treated with antibiotics were more often admitted to the hospital and had more often pulmonary infiltrates on chest radiograph than patients without antibiotics (Table 2).

TABLE 2.
TABLE 2.:
Baseline Characteristics RSV Patients with Viral Reference Standard

Subgroup Analysis

Of the 188 RSV patients 28 (47%), children required PICU admission. Of these, 17 (61%) had a bacterial coinfection and were treated with antibiotics. One RSV patient with a bacterial coinfection died due to respiratory and circulatory insufficiency. Of the patients with a viral infection admitted to the PICU, 8 (29%) received antibiotics. In the overall cohort, the percentage of bacterial coinfection was higher in children under 3 months of age (28.6%) than in children older than 3 months (12.4%), P = 0.034. Bacterial coinfection and PICU admissions were seen more often in the Dutch cohort than the Israeli cohort (25% vs. 7%, respectively, P < 0.001 and 34% vs. 1%, P < 0.001). The percentage of antibiotic overuse was similar in the Israeli and Dutch RSV cohort, 32% versus 27%, respectively (P = 0.52). When comparing the non-PICU cohort, the percentages of bacterial coinfections in the Dutch cohort decreases to 7.5% and 5.5% in the Israeli cohort (P = 0.74). The percentage of antibiotic overuse in this cohort is similar to the overall cohort.

DISCUSSION

In this study, we prospectively evaluated the frequency of bacterial coinfections in patients diagnosed with RSV LRTI and antibiotic overuse by using an expert panel as the reference standard. We showed that bacterial coinfections are rare with 14%, but that about one-third of all children with RSV LRTI were treated unnecessarily with antibiotics.

In general, viral LRTI is a risk factor for bacterial coinfection, with increasing severity of respiratory illness.18,19 The percentage of bacterial coinfections of 14% in our study is higher compared with other studies, most of them report bacterial coinfections up to 10%.3–10 The high proportion of bacterial coinfection is probably related to recruitment at our PICU. Previous research also showed that infants with severe bronchiolitis requiring admission on the intensive care unit have higher rates of bacterial coinfection.11,12,20–22 When selecting only the non-PICU cohort, the percentage of bacterial coinfections decreased to the more common percentage of 6%.

The prescription rate of antibiotics in patients with RSV in our cohort was low when compared with other studies. Patient recruitment was performed in both Israel and The Netherlands. Country-specific antibiotic prescription rates of the countries in which previous studies are performed, are higher in comparison with Israel and The Netherlands.3,9,23 Especially in The Netherlands, antibiotic consumption per person is relatively low.24 The percentage of antibiotic use on the PICU is also lower when compared with previous reported data on other PICU cohorts, but in line with 2 studies performed on other Dutch PICUs.11,12,20–22

Antibiotic overuse was similar in The Netherlands and Israel. Dutch and Israeli guidelines for management of pneumonia in children are similar and advise to start amoxicillin in children with clinical evidence of pneumonia.

A recent study showed that physicians are more likely to use antibiotics in non-RSV–infected patients compared with RSV.25 In our study, the nasal swabs were performed for research and were analyzed when patient recruitment was finished; therefore, the attending physician was not informed about the results of the nasal swab.

Strengths of this study deserve further discussion. First, a strength of the present study is the thorough nature of the reference standard. To our knowledge, the OPPORTUNITY and TAILORED studies are the first studies using an expert panel as the preferred reference standard to distinguish bacterial from viral infection to investigate the amount of antibiotic overuse in children with RSV LRTI. Clinical suspicion confirmed by microbiologic results is an approach often employed in other studies, but this method can be hampered by technical issues (e.g., culture contamination and uncertainty due to colonization) and microbiologic investigations were not performed in all patients, particularly in children with lower disease severity. Therefore, reported numbers of bacterial pneumonia in infants with RSV infections are mostly from studies performed on the PICU.11,12,20–22 Using an expert panel has the advantage of capturing a wider spectrum of illness severities in the resulting cohort, and therefore, is more likely to be generalizable to clinical practice.26,27 Second, we study the number of correct and incorrect antibiotic use in children with RSV infections. Other studies report on the number of bacterial coinfections separately from the antibiotic treatment numbers or excluded patients positive for bacterial pathogens.9,20,25 Therefore, it is in these studies not possible to calculate the amount of suspected antibiotic overuse. Limitations of our study should also be discussed. First, children below the age of 1 month were not included. Children younger than 30 days are at high risk of occult serious bacterial infections.28,29 A second limitation of the present study is that due to logistical reasons not all eligible children participated in this study, which may have introduced bias. Third, 4 (5%) of the RSV patients with a simple viral infection received antibiotics more than 72 hours after presentation. The expert panel was instructed to diagnose the patients at the moment of presentation. Therefore, we cannot exclude that these 4 patients developed a bacterial coinfection later on. Even when considering these antibiotic treatment as correct, still 28% of the RSV patients is treated unnecessary with antibiotics. Fourth, the expert panel was not able to diagnose all patients, although we found a high level of agreement (95%) between experts within the panels. Employing adjudication of an expert panel diagnosis is an approach commonly used in this area of research.27,30 Additionally, due to the low antibiotic prescribing rates in Israel and The Netherlands, the reported amount of antibiotic overuse is not generalizable to all other countries.23 The results of our study cannot be extrapolated to other viruses, the analysis of non-RSV patients was beyond the scope of this study. Finally, because of the relatively small number of patients with a bacterial coinfection, it was not possible to perform an accurate multivariate regression to define criteria for starting antibiotics in children infected with RSV. However, as described previously in other studies as well, the number of bacterial coinfections is low.

In conclusion, our data show that although bacterial coinfections in children with RSV infections are relatively rare, high percentages of antibiotic prescriptions are common. Further research is needed to develop accurate and practical tools to help physicians recognizing bacterial coinfections in children infected with RSV. In the meantime, a judicious approach to antimicrobial treatment is warranted for 1 of the most common diseases during early childhood.

ACKNOWLEDGMENTS

This publication has received funding from the EU’s Seventh Framework Programme FP7 under REA grant agreement No. HEALTH-F3-602860-2013 (TAILORED-Treatment; www.tailored-treatment.eu) and MeMed Diagnostics, Tirat Carmel, Israel.

REFERENCES

1. Hall CB, Weinberg GA, Iwane MK, et al.The burden of respiratory syncytial virus infection in young children. N Engl J Med. 2009;360:588598.
2. Farley R, Spurling GK, Eriksson L, et al.Antibiotics for bronchiolitis in children under two years of age. Cochrane Database Syst Rev. 2014;10:Cd005189.
3. Hall CB, Powell KR, Schnabel KC, et al.Risk of secondary bacterial infection in infants hospitalized with respiratory syncytial viral infection. J Pediatr. 1988;113:266271.
4. Hishiki H, Ishiwada N, Fukasawa C, et al.Incidence of bacterial coinfection with respiratory syncytial virus bronchopulmonary infection in pediatric inpatients. J Infect Chemother. 2011;17:8790.
5. Levine DA, Platt SL, Dayan PS, et alMulticenter RSV-SBI Study Group of the Pediatric Emergency Medicine Collaborative Research Committee of the American Academy of Pediatrics. Risk of serious bacterial infection in young febrile infants with respiratory syncytial virus infections. Pediatrics. 2004;113:17281734.
6. Titus MO, Wright SWPrevalence of serious bacterial infections in febrile infants with respiratory syncytial virus infection. Pediatrics. 2003;112:282284.
7. Purcell K, Fergie JConcurrent serious bacterial infections in 2396 infants and children hospitalized with respiratory syncytial virus lower respiratory tract infections. Arch Pediatr Adolesc Med. 2002;156:322324.
8. Bilavsky E, Shouval DS, Yarden-Bilavsky H, et al.A prospective study of the risk for serious bacterial infections in hospitalized febrile infants with or without bronchiolitis. Pediatr Infect Dis J. 2008;27:269270.
9. Bloomfield P, Dalton D, Karleka A, et al.Bacteraemia and antibiotic use in respiratory syncytial virus infections. Arch Dis Child. 2004;89:363367.
10. Cebey-López M, Pardo-Seco J, Gómez-Carballa A, et alGENDRES network. Bacteremia in children hospitalized with respiratory syncytial virus infection. PLoS One. 2016;11:e0146599.
11. Randolph AG, Reder L, Englund JARisk of bacterial infection in previously healthy respiratory syncytial virus-infected young children admitted to the intensive care unit. Pediatr Infect Dis J. 2004;23:990994.
12. van Woensel JB, von Rosenstiel IA, Kimpen JL, et al.Antibiotic use in pediatric intensive care patients with lower respiratory tract infection due to respiratory syncytial virus. Intensive Care Med. 2001;27:1436.
13. Diseases COoI. Antibiotic resistance threats in the United States 2013. Available at: http://www.cdc.gov/drugresistance/threat-report-2013/pdf/ar-threats-2013-508.pdf. Accessed 3 June 2015.
14. Laxminarayan R, Duse A, Wattal C, et al.Antibiotic resistance-the need for global solutions. Lancet Infect Dis. 2013;13:10571098.
15. van Houten CB, de Groot JAH, Klein A, et al.A host-protein based assay to differentiate between bacterial and viral infections in preschool children (OPPORTUNITY): a double-blind, multicentre, validation study. Lancet Infect Dis. 2017;17:431440.
16. Wacker C, Prkno A, Brunkhorst FM, et al.Procalcitonin as a diagnostic marker for sepsis: a systematic review and meta-analysis. Lancet Infect Dis. 2013;13:426435.
17. Rutjes AW, Reitsma JB, Coomarasamy A, et al.Evaluation of diagnostic tests when there is no gold standard. A review of methods. Health Technol Assess. 2007;11:iii, ix51.
18. Hament JM, Kimpen JL, Fleer A, et al.Respiratory viral infection predisposing for bacterial disease: a concise review. FEMS Immunol Med Microbiol. 1999;26:189195.
19. Hendaus MA, Jomha FA, Alhammadi AHVirus-induced secondary bacterial infection: a concise review. Ther Clin Risk Manag. 2015;11:12651271.
20. Thorburn K, Harigopal S, Reddy V, et al.High incidence of pulmonary bacterial co-infection in children with severe respiratory syncytial virus (RSV) bronchiolitis. Thorax. 2006;61:611615.
21. Kneyber MC, Blussé van Oud-Alblas H, van Vliet M, et al.Concurrent bacterial infection and prolonged mechanical ventilation in infants with respiratory syncytial virus lower respiratory tract disease. Intensive Care Med. 2005;31:680685.
22. Duttweiler L, Nadal D, Frey BPulmonary and systemic bacterial co-infections in severe RSV bronchiolitis. Arch Dis Child. 2004;89:11551157.
23. The center for disease dynamics EP. Antibiotic prescribing rates by country. Available at: http://www.cddep.org/tool/antibiotic_prescribing_rates_country#sthash.kiRPO6aH.dpbs. Accessed 02 Feb 2018.
24. van de Sande-Bruinsma N, Grundmann H, Verloo D, et al.Antimicrobial drug use and resistance in Europe. Emerg Infect Dis. 2008;14:17221730.
25. Goriacko P, Saiman L, Zachariah PAntibiotic use in hospitalized children with respiratory viruses detected by multiplex polymerase chain reaction. Pediatr Infect Dis J. 2017.
26. Reitsma JB, Rutjes AW, Khan KS, et al.A review of solutions for diagnostic accuracy studies with an imperfect or missing reference standard. J Clin Epidemiol. 2009;62:797806.
27. Bertens LC, Broekhuizen BD, Naaktgeboren CA, et al.Use of expert panels to define the reference standard in diagnostic research: a systematic review of published methods and reporting. PLoS Med. 2013;10:e1001531.
28. Bachur RG, Harper MBPredictive model for serious bacterial infections among infants younger than 3 months of age. Pediatrics. 2001;108:311316.
29. Milcent K, Faesch S, Gras-Le Guen C, et al.Use of procalcitonin assays to predict serious bacterial infection in young febrile infants. JAMA Pediatr. 2016;170:6269.
30. Oved K, Cohen A, Boico O, et al.A novel host-proteome signature for distinguishing between acute bacterial and viral infections. PLoS One. 2015;10:e0120012.
Keywords:

RSV; respiratory tract infection; antibiotic use; pulmonology; infectious diseases

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.