Abstract
We investigated the concordance between the Unyvero Hospitalized Pneumonia (HPN) application and quantitative culture for detection of bacterial pathogens from serial lower respiratory tract (LRT) specimens collected from the same subject. Comparison of results from HPN application and culture was evaluated using 69 LRT samples from 27 subjects, using two evaluation approaches. False positive detections by the HPN application was 29% (20/69) in Evaluation I vs 10% (7/68) in Evaluation II. Additional pathogens detected by the HPN application could be confirmed in many instances by culture positivity for the same organism from previous or subsequent samples from the same subject.
Avoid common mistakes on your manuscript.
Introduction
Since the introduction of multiplex PCR-systems for diagnosis of pneumonia, numerous comparative studies to the reference methodology — quantitative respiratory culture — have been performed [1,2,3]. Results from these studies indicate that the PCR-based diagnostic systems can provide rapid results for both pathogen identification and resistance markers as compared to the microbiological culture-based diagnosis. Furthermore, some of these studies have demonstrated that the PCR-based systems can detect additional pathogens that were not reported by culture. Interpretation of the results with regard to these additional pathogens detected by the PCR-based systems can often be challenging in the absence of culture positivity. Consequently, low positive predictive values (PPVs) of the PCR-based systems, ranging from 46.9 to 78.6% were reported in the published literature, when the comparisons were done assuming that the conventional microbiologic techniques are 100% sensitive and specific [3, 4]. The performance characteristics of the Unyvero Hospitalized Pneumonia (HPN) application (Curetis GmbH, Germany) in comparison with standard-of-care microbiological culture for detection of bacterial pathogens from lower respiratory tract (LRT) samples obtained from critically ill COVID-19 patients were recently reported [3].
Herein, we present a follow-up study aimed to investigate the concordant and discrepant results comparing the Unyvero HPN and culture results for detection of microorganisms from serial specimens collected from the same subject.
Material and methods
A laboratory-based evaluation study was undertaken at the clinical microbiology laboratory of Karolinska University Hospital (KUH), Solna, Stockholm, Sweden. Frozen lower respiratory tract (LRT) samples from the biobank of the study laboratory were tested with the Unyvero HPN application. All samples had been subjected to standard of care (SoC) testing previously. Detailed description of the methods used for testing the samples using the SoC and the HPN application was described in detail previously [3].
Clinical samples and data
Samples for the present evaluation were selected from the laboratory information system (LIS) for patients who met the following criteria: (1) samples were obtained from subjects admitted in the intensive care unit with COVID-19, (2) samples were collected from adult patients, (3) at least two or more samples collected on separate days were available from an individual subject, and (4) samples from patients with at least one episode of culture positivity for an HPN panel organism. Sample types, sample collection dates, and microbiology culture findings of the study samples and baseline demographic data of the study subjects were retrospectively collected from the LIS. Clinical data of the study subjects were retrospectively collected from electronic health records by the study physicians.
Definitions used for interpretation of the HPN application results
Full concordance (FC)
Full concordance was considered when the results from the HPN application were identical for the presence/absence of one or more pathogens by SoC: samples with SoC + /HPN + or SoC − /HPN − for a given pathogen; normal microbiota on SoC/HPN − ; only off-panel pathogen SoC + / HPN − .
Concordance by correlation (CC)
Concordance by correlation was considered when HPN application was positive for a pathogen that was negative by SoC from the same sample but was positive by culture in a previous or a subsequent sample from the same patient within ± 7 days.
Partial concordance (PC)
Partial concordance was considered when HPN application detected the same pathogens that were detected by SoC plus additional pathogen(s) that were not detected by SoC and failed achieving concordance by correlation.
Discordance (D)
Discordance was considered when an on-panel pathogen was culture positive but was not detected by the HPN application (SoC + /HPN −). Discordant results were also considered when a sample had only one pathogen detected by the HPN application but was not detected by SoC (SoC − /HPN +).
Concordance and discordance (C/D)
There were samples that were positive for more than one pathogen by HPN application but were not detected by SoC. One pathogen was concordant by correlation by culture in a previous or a subsequent sample from the same patient with in ± 7 days; the other pathogen was not detected by SoC.
Data analysis
The first evaluation (Evaluation I) was performed in a typical manner for a comparative study where Unyvero results were compared with culture results per subject. The second evaluation (Evaluation II) explored interpretation of Unyvero false-positive results which were corroborated by culture from a different sample taken at a later or at an earlier time point from the same subject. In this approach, all Unyvero false positive results were considered true positives, if culture confirmed this result for any other sample (collected at an earlier or later time point) from the same subject.
Results
Baseline characteristics of the study samples:
Sixty-nine samples obtained from 27 subjects (15 patients with two, 9 patients with three, and three patients with four samples collected at subsequent time point per subject) were included. Collection of the first respiratory sample for microbiological investigations before administration of an antibiotic was done in 15/27 (55.5%) subjects. Clinical characteristics of the study subjects are listed in Table S1.
Performance of the HPN application
Full concordance of results from both methods was observed among 45/69 (65.2%) samples (Fig. 1). Nine of the 27 patients with 22 samples (Table S2) showed 100% fully concordant results for all samples collected. Concordance by correlation (CC) was observed among 12/69 (17.3%) samples. At least one episode of CC could be achieved in 11/27 patients (Tables 1 and 2).CC could be achieved from a previous sample and a subsequent sample collected in six subjects each. Discordant results were observed among 7/69 (10%) samples. Among these, SoC + /HPN − result for a pathogen was observed among five samples (Table 2). Among these five samples, the HPN application failed at detecting a pathogen in four samples and misidentified the species name in one sample (Subject 23). Among the four samples with SoC + /HPN − results, the SoC results for three samples were above the lower detection limit of the HPN application, and in one sample (sample 2 from subject 16), the culture positivity was below the detection limit of the HPN application. Discordance, in terms of SoC − /HPN + for additional pathogens was observed among two samples (sample 2 from subject 4; sample 3 from subject 23, both listed in Table 1). There were partial concordant results where the Unyvero application detected additional pathogens in 4/69 samples derived from three patients (Table 1) in addition to the fully matched pathogens. These include two Staphylococcus aureus and two Serratia marcescens. Finally, concordance and discordance (C/D) was observed in one sample (sample 1 from the subject 27, listed in Table 1).
Comparison of the results from HPN application with SoC testing
Evaluation I
When the results from the HPN application were compared with the culture findings per subject time point, the HPN application generated false positive results in 20/69 (29%) samples (Table 3).
Evaluation II
When the results from the HPN application were compared with culture findings per subject time point along with correlation with a previous or a subsequent culture finding, false positive results were observed in 7/68 (10%) samples (Table 4).
Discussion
In this study, the false positive results generated from the Unyvero HPN application could be reduced from 29 to 10% by correlating with the culture findings from a previous or a later respiratory sample. Detection of additional pathogens from a given respiratory sample using rapid molecular diagnostic tests has been a consistent observation in several studies when the results from these tests were directly compared with the culture findings [5, 6], and the approach to use a ± 7-day time-window, as in the present study, could be a rational way to validate new diagnostic tests.
At least one additional pathogen was detected from 20/69 (29%) of the samples tested, and the additional pathogens detected in 12/20 (60%) samples were detected by culture from a previous or a subsequent sample collected (± 7 days) and tested from the same subject in six samples each. This observation underscores the ability of the HPN application in detecting a potential pneumonia pathogen earlier than culture and/or very early during infection. Additional pathogens detected from some of the study samples could not be correlated with culture positivity for the same pathogen from previous or subsequent samples, and results from these tests were considered as discordant or partial concordant (Tables 1 and 2). The exact reasons for detection of additional pathogens from these samples could not be determined in the present study. However, it is possible that some of the microorganisms detected were colonizers present in low amounts below the current diagnostic thresholds for reporting. Another important factor that could have had led to additional detection (SoC − / HPN +) for a given organism is the antibiotic exposure before sample collection. The Unyvero HPN application detected an additional pathogen from at least one sample collected from 14/27 (52%) subjects in the present study. Among these 14 subjects, 10 of them had exposure to antibiotics before the collection of samples for culture, and it is plausible that the exposure of the antibiotic had a negative influence on the yield of the cultures from these samples.
In conclusion, the additional pathogens detected by the Unyvero HPN application from a given LRT sample could be confirmed in many instances by culture positivity for the same microorganism from a previous or a subsequent sample obtained from the same subject.
Data availability
All relevant data are available in the manuscript.
References
Pulido MR, Moreno-Martínez P, González-Galán V, Fernández Cuenca F, Pascual Á, Garnacho-Montero J et al (2018) Application of BioFire FilmArray Blood Culture Identification panel for rapid identification of the causative agents of ventilator-associated pneumonia. Clin Microbiol Infect 24(1213):e1-1213.e4
Mitton B, Rule R, Said M (2021) Laboratory evaluation of the BioFire FilmArray Pneumonia plus panel compared to conventional methods for the identification of bacteria in lower respiratory tract specimens: a prospective cross-sectional study from South Africa. Diagn Microbiol Infect Dis 99:115236
Tellapragada C, Giske CG (2021) The Unyvero Hospital-Acquired pneumonia panel for diagnosis of secondary bacterial pneumonia in COVID-19 patients. Eur J Clin Microbiol Infect Dis 40:2479–2485
Foschi C, Zignoli A, Gaibani P, Vocale C, Rossini G, Lafratta S et al (2021) Respiratory bacterial co-infections in intensive care unit-hospitalized COVID-19 patients: conventional culture vs BioFire FilmArray pneumonia Plus panel. J Microbiol Methods 186:106259
Enne VI, Aydin A, Baldan R, Owen DR, Richardson H, Ricciardi F, et al (2022) Multicentre evaluation of two multiplex PCR platforms for the rapid microbiological investigation of nosocomial pneumonia in UK ICUs: the INHALE WP1 study. Thorax: thoraxjnl-2021–216990.
Murphy CN, Fowler R, Balada-Llasat JM, Carroll A, Stone H, Akerele O et al (2020) Multicenter evaluation of the BioFire FilmArray pneumonia/pneumonia plus panel for detection and quantification of agents of lower respiratory tract infection. J Clin Microbiol 58:e00128-e220
Acknowledgements
The authors thank Curetis, Germany GmbH, for kindly providing the HPN cartridges for the present evaluation.
Funding
Open access funding provided by Karolinska Institute.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Ethics approval
Ethical permit was obtained from the Swedish Ethics Review Authority (Ref No:2020–04999).
Consent to participate
Not applicable.
Consent for publication
Not applicable.
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Tellapragada, C., Ydsten, K.A., Ternhag, A. et al. Evaluation of a pneumonia multiplex PCR panel for detection of bacterial respiratory tract pathogens from serial specimens collected from hospitalized COVID-19 patients. Eur J Clin Microbiol Infect Dis 41, 1093–1098 (2022). https://doi.org/10.1007/s10096-022-04466-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10096-022-04466-9