Comparison of microbial composition of cough swabs and sputum for pathogen detection in patients with cystic fibrosis

Background: With the continued advancement of CFTR modulator therapies there is likely to be a burgeoning population of adult cystic fibrosis (CF) patients unable to expectorate sputum. Consequently, the detection and surveillance of pulmonary colonisation, previously reliant on sputum culture, needs re-examining. We hypothesised that cough swabs analysed with culture-independent analysis of the 16S gene could serve as a surrogate for colonisation of the lower airways. Methods: Cough swabs and sputum samples were prospectively collected from consecutive adults and children with CF across two sites at regular outpatient appointments. Conventional culture analysis and next generation sequencing were used to compare paired same day samples. Results: Twenty-two adults and 8 paediatric patients provided 75 paired cough swabs and sputum samples. Alpha diversity measures showed increased bacterial richness in sputum, while evenness and Simpson's diveristy index were higher in cough swabs. Within each sampling technique, microbial composition showed greater similarity when considering intra-patient variation. Poor concordance was observed between culture independent cough swabs and culture dependent/independent sputum analysis for specific pathogens, with cough swabs unable to accurately identify commonly associated CF pathogens (AUROCC range: 0.51 to 0.64). Conclusion: Culture independent analysis of cough swabs provides an inaccurate diagnosis of lower respiratory tract colonisation and should not be used as a diagnostic test in patients with CF.