https://orcid.org/0000-0003-1706-6869
Figures
Abstract
Introduction
Using respiratory virus rapid diagnostic tests in the emergency department could allow better and faster clinical management. Point-of-care PCR instruments now provide results in less than 30 minutes. The objective of this study was to assess the impact of the use of a rapid molecular diagnostic test, the cobas® Influenza A/B & RSV Assay, during the clinical management of emergency department patients.
Methods
Patients (adults and children) requiring admission or suffering from an underlying condition at risk of respiratory complications were prospectively recruited in the emergency department of four hospitals in the Brussels region. Physicians’ intentions regarding admission, isolation, antibiotic, and antiviral use were collected before and after performing the rapid molecular test. Additionally, a comparison of the analytical performance of this test against antigen rapid tests and viral culture was performed as well as a time-to-result evaluation.
Results
Among the 293 patients recruited, 90 had a positive PCR, whereas 44 had a positive antigen test. PCR yielded a sensitivity of 100% for all targets. Antigen tests yielded sensitivities ranging from 66.7% for influenza B to 83.3% for respiratory syncytial virus (RSV). The use of PCR allowed a decrease in the overall need for isolation and treatment by limiting the isolation of negative patients and antibiotic use for positive patients. Meanwhile, antiviral treatments better targeted patients with a positive influenza PCR.
Conclusion
The use of a rapid influenza and RSV molecular test improves the clinical management of patients admitted to the emergency department by providing a fast and reliable result. Their additional cost compared to antigen tests should be balanced with the benefit of their analytical performance, leading to efficient reductions in the need for isolation and antibiotic use.
Citation: Yin N, Van Nuffelen M, Bartiaux M, Préseau T, Roggen I, Delaunoy S, et al. (2022) Clinical impact of the rapid molecular detection of RSV and influenza A and B viruses in the emergency department. PLoS ONE 17(9): e0274222. https://doi.org/10.1371/journal.pone.0274222
Editor: Baochuan Lin, Defense Threat Reduction Agency, UNITED STATES
Received: March 16, 2022; Accepted: August 23, 2022; Published: September 2, 2022
Copyright: © 2022 Yin et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: NO - Roche diagnostics (https://diagnostics.roche.com/) supplied instruments and reagents needed for this study. Roche diagnostics did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. No personal grants or funding was perceived by the authors for this study.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The COVID-19 pandemic has highlighted the need for rapid and trustworthy diagnostic tests for respiratory tract infections to assess patients’ potential infectiousness and to set containment measures [1, 2]. Various studies have previously demonstrated that a fast and reliable diagnosis of respiratory infections improves the clinical management during seasonal epidemics of respiratory syncytial virus (RSV) and influenza A and B viruses [3, 4]. In Europe, before the COVID-19 pandemic, universal health coverage rarely assured access to molecular diagnostic testing for these infections, limiting their prescription in favor of less expensive antigen tests. In Belgium, for instance, SARS-CoV-2 PCR detection is now reimbursed, but for RSV and influenza A and B viruses, only antigen detection is covered [5]. However, compared to RT‒PCR, the sensitivity of these antigen rapid diagnostic tests was estimated to be as low as 53.9% in adults and 64.6% in children regarding influenza A [6] and 74% in children regarding RSV [7]. Furthermore, clinical judgment, followed by PCR or point-of-care testing (POCT), was found to be cost-effective in a setting where influenza probability was high [8]. Fast “sample-in, result-out” PCR instruments, such as the Roche cobas® Liat®, are now providing point-of-care PCR results in less than 30 minutes (a run on the instrument is completed in 21 minutes once the sample is loaded; the hands-on time is approximately 5 minutes) [9]. The objective of this study was to assess the impact of the use of the cobas® Liat® Influenza A/B & RSV assay on the clinical management of emergency department (ED) patients requiring admission or presenting an underlying condition at risk of respiratory complications.
Material and methods
Population and data collection
Patients attending the EDs of four hospitals located in the Brussels area of Belgium were prospectively recruited. To be included, patients needed to meet at least one of the following inclusion criteria: (i) respiratory symptoms and a pretest indication of hospitalization by the physician; (ii) respiratory symptoms and an underlying condition at risk of respiratory complication following influenza infection (as described by the European Centre for Disease Prevention and Control and the US Centers for Disease Control: age ≥ 65 years old or < 2 years old, pregnancy, chronic medical conditions, immunocompromised patient, etc. [10, 11]); (iii) age < 3 months with unexplained fever. This study did not use any exclusion criteria. When including the patient and prescribing the test, physicians indicated their intentions regarding antimicrobial treatment, isolation, and hospitalization by answering a questionnaire. The same questionnaire was completed again once the PCR results were available to evaluate the posttest intentions. Written informed consent was collected from each participant or their guardian. The ethics committee of each hospital approved the study.
Specimen collection and analyses
The accepted samples were nasopharyngeal swabs in 3 mL of universal transport medium (UTM). Nasopharyngeal aspirates diluted with 3 mL of UTM were also accepted for children. Approximately 200 μL was immediately used to perform a rapid PCR using the cobas® Liat® Influenza A/B & RSV assay (Roche Diagnostics, Indianapolis, IN) following the manufacturer’s instructions in a 24/7 on-site laboratory or in a point-of-care setting. PCR results were instantly transmitted to the result servers. For PCRs performed at on-site laboratories, PCR time-to-result was defined as the time between the prescription and the availability of the results. In addition, antigen rapid diagnostic tests (RDTs) were performed after PCR using the Influ A+B K-Set (Coris BioConcept, Gembloux, Belgium) for influenza A and influenza B detection and RSV K-SeT (Coris BioConcept) for RSV detection on the same sample (200 μL/test). Upon arrival in the virology laboratory, 1 mL of UTM was also inoculated onto confluent Vero (African green monkey kidney), MRC5 (human lung) and LLC-MK2 (rhesus monkey kidney) cell cultures (Vircell, Granada, Spain) in 24-well or 6-well tissue culture plates (Greiner-Bio One, Vilvoorde, Belgium). Cultures were incubated at 36°C in a 5% CO2 atmosphere for 2 weeks for Vero and LLC-MK2 cells and 3 weeks for MRC5 cells. The media were replaced weekly. Cultures were examined every two to three days using an inverted microscope. Hemadsorption was performed on LLC-MK2 cells at the end of the second week of incubation.
Statistical analysis
To assess the analytical performance of a molecular detection technique against culture and antigen detection, a composite reference standard was constructed as recommended [12]. Samples considered positive for a viral pathogen were defined as those testing positive for this viral pathogen by at least 2 of the 3 techniques used and negative as those that tested negative by at least 2 of the 3 techniques. Statistical analyses were performed using Analyse-it® for Microsoft Excel v5.30.4 (Analyse-it Software, Leeds, United Kingdom). Proportion variations between pre-test and post-test clinical intentions were evaluated using the McNemar-Mosteller exact test.
Results
Two hundred and ninety-three patients were recruited, including 68 children (< 15 years old) (Table 1), from February to March 2020. Among them, 71 (incl. 25 children) had a positive rapid PCR result for influenza A only, 10 (incl. 2 children) for influenza B only, 1 child for both influenza A and B, 8 (incl. 5 children) for RSV and 203 (incl. 35 children) were negative for all targets. The positive agreement between PCR and RDT ranged from 36.3% (4/11) for influenza B to 62.5% (5/8) for RSV. One sample was positive for influenza A using RDT although negative by culture and PCR. Likewise, one sample was positive for influenza A using culture although negative by PCR and RDT. Thus, using the composite reference standard, PCR reached a sensitivity of 100% for the 3 viruses targeted (Table 2).
Clinical intentions regarding admission were not influenced by the result of the test (Table 3). However, an overall decrease in the intentions of admission in isolation (odds ratio (OR) = 0.381; 95% confidence interval (CI) = 0.200–0.692)), antibiotic use (OR = 0.486; CI = 0.261–0.876) and oseltamivir use (OR = 0.543; CI = 0.320–0.903) was observed. These trends were the result of both a sharp decrease in isolation intentions for people with a negative PCR (OR = 0.077; CI = 0.015–0.242) and an increase in isolation intentions for people with a positive PCR (OR = 4.333; CI = 1.191–23.707). Similarly, patients with a positive PCR were less likely to be prescribed antibiotics (OR = 0.235; CI = 0.058–0.721), and oseltamivir treatment was more specifically prescribed for patients with an influenza-positive PCR (OR = 12.500; CI = 3.117–108.889), including for patients with a positive PCR but negative RDT results (OR = 8.500; CI = 2.018–75.851). Positive RDT alone would not have resulted in a significant increase in isolation (OR = +∞; CI = 0.660 –+∞).
Two hundred and eighty-four (97.3%) samples were analyzed in a 24/7 on-site laboratory, and the remaining samples were analyzed in the ED as a point of care. The median time-to-result of the PCR in the laboratory was 60 minutes (CI = 53–76 minutes), with a 90th percentile of 164 minutes.
Discussion
Previous studies have underscored the impact of the rapid detection of influenza and RSV in the management of patients [13–16]. We showed that an overall significant decrease in antibiotic prescription could be obtained by providing a relevant diagnostic tool combining speed and sensitivity. The excellent sensitivity of the cobas® Liat® assay allowed an optimized assessment of the need for isolation for the inpatients, ensuring proper isolation for almost all infected patients, which likely led to a reduction in nosocomial transmission. Indeed, less than half of the patients with a positive PCR also had a positive RDT, emphasizing the importance of using a molecular test efficiently to improve antibiotic and antiviral stewardship while offering a real assessment for the need for costly hospitalization in isolation. The median days since symptom onset was 2 for RDT-positive patients, whereas it was 2.5 for PCR-positive patients, although the difference was not statistically significant. RDTs are well known to be more sensitive for higher viral loads and hence more infectious patients. Nevertheless, recruited patients were symptomatic; thus, better performance is expected from clinicians for a positive diagnosis regardless of the number of days since symptom onset. The overall decrease in the isolation indications for patients with a negative test should be balanced with the clinical presentation of the patient to prevent the nosocomial transmission of other respiratory pathogens. Nevertheless, in a pandemic context where isolation resources can be scarce, the use of a negative reliable test to decrease the need for isolation should be considered.
In our study, the median time of response from the laboratory was 60 minutes for a less than 30-minute assay. Nevertheless, the time-to-result in this setting is dependent on the overall workload in the on-site laboratory. Using this assay as a POCT at the ED would certainly decrease the time-to-result while increasing the autonomy of the ED team regarding the general management of the workflow and the real-time adaptation of the indications, depending on the current level of bed occupancy and the epidemic situation.
However, such a strategy has a cost and should be further examined through a medico-economic study to better determine the optimal target population while taking into account the subsequent decrease in nosocomial outbreaks and the targeted use of individual protection equipment and antibiotics. In a Dutch study, the daily direct cost of isolating patients ranged from €28 (£23/$31) to €41 (£34/$46) [17]. At the time of writing, the price of the cobas® Liat® Influenza A/B & RSV was 39€/test (£33/$44) (excluding instruments and workforce). In Belgium, an antigen test is reimbursed 8.06€ (£6.74/$9.00) per target (maximum 3) and a viral culture, 45.17€ (£37.75/$50.48). Decreasing the number of incorrectly isolated patients would likely balance the costs of performing PCR instead of RDTs and viral cultures for inpatients during seasonal epidemics.
Even as we were able to evaluate the cobas® Liat® Influenza A/B & RSV assay in the laboratory, this study was prematurely interrupted by the surge of the COVID-19 pandemic in Belgium in March 2020. This event prevented us from evaluating the benefit of the use of this assay as a POCT because this assay was not relevant for the epidemic at the time. The need for human and technical resources also prevented us from cross checking PCR-only positive tests with another molecular method, likely underestimating the specificity of this assay. Likewise, we used a composite reference standard to balance the well-known lower sensitivity of RDT and culture. In the beginning of 2020, the COVID-19 pandemic was associated with smaller outbreaks of influenza and RSV. The positivity rate was lower than that in a previous study [9] and could be an additional limitation. Following the emergence of COVID-19, the implementation of the cobas® Liat® as a POCT in the ED raised biosafety concerns. This assay requires pipetting UTM to the reagent tube, which can be considered a risk of projection and aerosolization. To solve this, the operator performing the sampling wearing personal protection equipment loaded the reagent tube with UTM at the bed side. The sealed tube was then transported to the instrument located in a separate room [18].
In conclusion, using a rapid molecular assay, such as the cobas® Liat® Influenza A/B & RSV, improves the clinical management of patients by refining the indications of isolation, antibiotic, and antiviral treatment due to better analytical performances than RDTs. Using this instrument as a POCT in the ED could provide an efficient 24/7 solution to dispatch inpatients and quickly and optimally manage at-risk outpatients.
Acknowledgments
We wish to thank the nurses of the emergency department of our hospitals for their amazing work and their collaboration in this study. We would like to thank Dustin DeMeglio for his help in English writing.
References
- 1. Vandenberg O, Martiny D, Rochas O, van Belkum A, Kozlakidis Z. Considerations for diagnostic COVID-19 tests. Nat Rev Microbiol. 2021 Mar;19(3):171–83. pmid:33057203
- 2. Sheridan C. Coronavirus testing finally gathers speed. Nat Biotechnol [Internet]. 2020 Nov 5 [cited 2020 Dec 20]; Available from: http://www.nature.com/articles/d41587-020-00021-z pmid:33154578
- 3. Maltezou HC. Nosocomial influenza: new concepts and practice: Curr Opin Infect Dis. 2008 Aug;21(4):337–43. pmid:18594283
- 4. Bont L. Nosocomial RSV infection control and outbreak management. Paediatr Respir Rev. 2009 Jun;10:16–7. pmid:19651394
- 5.
Nomenclature des prestations de santé [Internet]. Sect. Biologie Clinique—Article 24. Available from: https://www.inami.fgov.be/SiteCollectionDocuments/nomenclatureart24_20210701_01.pdf
- 6. Chartrand C, Leeflang MMG, Minion J, Brewer T, Pai M. Accuracy of Rapid Influenza Diagnostic Tests: A Meta-analysis. Ann Intern Med. 2012 Apr 3;156(7):500. pmid:22371850
- 7. 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. Tang Y-W, editor. J Clin Microbiol. 2015 Dec;53(12):3738–49. pmid:26354816
- 8. Lee BY, McGlone SM, Bailey RR, Wiringa AE, Zimmer SM, Smith KJ, et al. To Test or to Treat? An Analysis of Influenza Testing and Antiviral Treatment Strategies Using Economic Computer Modeling. Galvani AP, editor. PLoS ONE. 2010 Jun 23;5(6):e11284. pmid:20585642
- 9. Gibson J, Schechter-Perkins EM, Mitchell P, Mace S, Tian Y, Williams K, et al. Multi-center evaluation of the cobas ® Liat ® Influenza A/B & RSV assay for rapid point of care diagnosis. J Clin Virol. 2017 Oct;95:5–9.
- 10.
European Centre for Disease Prevention and Control. Factsheet about seasonal influenza [Internet]. [cited 2021 Jul 27]. Available from: https://www.ecdc.europa.eu/en/seasonal-influenza/facts/factsheet
- 11.
Centers for Disease Control and Prevention. Influenza Antiviral Medications: Summary for Clinicians [Internet]. [cited 2021 Jul 27]. Available from: https://www.cdc.gov/flu/professionals/antivirals/summary-clinicians.htm
- 12. Banoo S, Bell D, Bossuyt P, Herring A, Mabey D, Poole F, et al. Evaluation of diagnostic tests for infectious diseases: general principles. Nat Rev Microbiol. 2010 Dec;8(S12):S16–28. pmid:21548184
- 13. Bonner AB, Monroe KW, Talley LI, Klasner AE, Kimberlin DW. Impact of the Rapid Diagnosis of Influenza on Physician Decision-Making and Patient Management in the Pediatric Emergency Department: Results of a Randomized, Prospective, Controlled Trial. PEDIATRICS. 2003 Aug 1;112(2):363–7. pmid:12897288
- 14. Rahamat‐Langendoen J, Groenewoud H, Kuijpers J, Melchers WJG, Wilt GJ. Impact of molecular point‐of‐care testing on clinical management and in‐hospital costs of patients suspected of influenza or RSV infection: a modeling study. J Med Virol. 2019 Aug;91(8):1408–14. pmid:30950066
- 15. Falsey AR, Murata Y, Walsh EE. Impact of rapid diagnosis on management of adults hospitalized with influenza. Arch Intern Med. 2007 Feb 26;167(4):354–60. pmid:17242309
- 16. Busson L, Bartiaux M, Brahim S, Konopnicki D, Dauby N, Gérard M, et al. Contribution of the FilmArray Respiratory Panel in the management of adult and pediatric patients attending the emergency room during 2015–2016 influenza epidemics: An interventional study. Int J Infect Dis. 2019 Jun;83:32–9. pmid:30926539
- 17. van Dijk MD, Voor in ‘t holt AF, Polinder S, Severin JA, Vos MC. The daily direct costs of isolating patients identified with highly resistant micro-organisms in a non-outbreak setting. J Hosp Infect. 2021 Mar;109:88–95. pmid:33359899
- 18. Yin N, Debuysschere C, Decroly M, Bouazza F-Z, Collot V, Martin C, et al. SARS-CoV-2 Diagnostic Tests: Algorithm and Field Evaluation From the Near Patient Testing to the Automated Diagnostic Platform. Front Med. 2021;8:650581.