Diagnostic utility of oropharyngeal swabs as an alternative to lower respiratory tract samples for PCR-based syndromic testing in patients with community-acquired pneumonia

Abstract

Syndromic PCR-based analysis of lower respiratory tract (LRT) samples in patients with community-acquired pneumonia (CAP) improves the bacterial yield and time-to-results compared to culture-based methods. However, obtaining adequate sputum samples can be challenging and is frequently not prioritized in the emergency department (ED). In this study, we assess the concordance of microbiological detections between oropharyngeal- (OP) and LRT samples from patients presenting to the ED with CAP using a syndromic PCR-based respiratory panel [Biofire FilmArray Pneumonia plus (FAP plus)]. Paired OP- and high-quality LRT samples were collected from 103 patients with confirmed CAP, who had been included in a randomized controlled trial (NCT04660084) or a subsequent observational study at Haukeland University Hospital, and analyzed using the FAP plus. The LRT samples were obtained mainly by sputum induction (88%). Using the LRT samples as a reference standard, the positive percent agreement (PPA), negative percent agreement (NPA), and overall percent agreement for the most common bacterial pathogens in CAP, Streptococcus pneumoniae and Haemophilus influenzae, were 85%, 99% and 95%, and 86%, 98% and 93%, respectively. For Moraxella catarrhalis, the PPA was lower (74%), while the NPA was 100%. For bacteria that are less likely causes of uncomplicated CAP (e.g., Staphylococcus aureus and Enterobacterales) the results were more divergent. In conclusion, the FAP plus detects the most common CAP pathogens S. pneumoniae and H. influenzae from OP samples with high PPAs and excellent NPAs when compared with LRT samples. For these pathogens, the PPAs for OP samples were higher than previous reports for nasopharyngeal samples. This suggests that analysis of OP samples with syndromic PCR panels could represent an alternative approach for rapid microbiological testing in the ED, especially in patients where LRT samples are difficult to obtain. Divergent results for bacteria that are less likely to cause uncomplicated CAP do, however, emphasize the need for clinical evaluation of positive test results.

Keywords: Biofire FilmArray Pneumonia panel; Haemophilus influenzae; Streptococcus pneumoniae; community-acquired pneumonia; molecular diagnostics; oropharyngeal swab; sputum; syndromic testing; upper respiratory tract sample.

Fig 1

Study flowchart. Abbreviations: CAP, community-acquired pneumonia; LRT, lower respiratory tract; OP, oropharyngeal.

Fig 2

Detection rates by the Biofire FilmArray Pneumonia plus panel in paired OP and LRT samples from patients with CAP. Stratified by type of specimen. ^a Acinetobacter calcoaceticus-baumannii complex. ^b Enterobacter cloacae complex, Escherichia coli, Klebsiella aerogenes, Klebsiella pneumoniae, Klebsiella oxytoca, Serratia marcescens, Proteus species. ^c Adenovirus, coronavirus (229E, OC43, HKU1, and NL63), parainfluenza virus, rhino-/enterovirus, RS virus, human metapneumovirus, influenza virus. Abbreviations: OP, oropharyngeal; LRT, lower respiratory tract; CAP, community-acquired pneumonia.